kitten experiment psychology

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Front Psychol

Rediscovering Richard Held: Activity and Passivity in Perceptual Learning

Fernando bermejo.

1 Centro de Investigación y Transferencia en Acústica, Universidad Tecnológica Nacional − Facultad Regional Córdoba, CONICET, Córdoba, Argentina

2 Facultad de Psicología, Universidad Nacional de Córdoba, Córdoba, Argentina

3 Consejo Nacional de Investigaciones Científicas y Tecnológicas, Buenos Aires, Argentina

Mercedes X. Hüg

Ezequiel a. di paolo.

4 Ikerbasque, Basque Foundation for Science, Bilbao, Spain

5 IAS Research Center for Life, Mind and Society, University of the Basque Country, San Sebastián, Spain

6 Centre for Computational Neuroscience and Robotics, University of Sussex, Brighton, United Kingdom

Understanding the role of self-generated movements in perceptual learning is central to action-based theories of perception. Pioneering work on sensory adaptation by Richard M. Held during the 1950s and 1960s can still shed light on this question. In a variety of rich experiments Held and his team demonstrated the need for self-generated movements in sensory adaptation and perceptual learning. This body of work received different critical interpretations, was then forgotten for some time, and saw a surge of revived interest within embodied cognitive science. Through a brief review of Held’s work and reactions to it, we seek to contribute to discussions on the role of activity and passivity in perceptual learning. We classify different positions according to whether this role is considered to be contextual (facilitatory, but not necessary), enabling (causally necessary), or constitutive (an inextricable part of the learning process itself). We also offer a critique of the notions of activity and passivity and how they are operationalized in experimental studies. The active-passive distinction is not a binary but involves a series of dimensions and relative degrees that can make it difficult to interpret and replicate experimental results. We introduce three of these dimensions drawing on work on the sense of agency: action initiation, control, and monitoring. These refinements in terms of causal relations and dimensions of activity-passivity should help illuminate open questions concerning the role of activity in perception and perceptual learning and clarify the convergences and differences between enaction and ecological psychology.

Introduction

Action-based accounts of perception maintain that there are functional and conceptual links between action and perception (e.g., O’Regan and Noë, 2001 ; Noë, 2004 ; Pulvermüller and Fadiga, 2010 ). These perspectives are advocated both by enactivists and ecological psychologists (e.g., Varela et al., 1991 ; Reed, 1996 ; Chemero, 2009 ; Di Paolo et al., 2017 ) and can serve to highlight the convergences and the differences between these approaches. In the extensive literature on the subject one experiment has become iconic. In a study conducted in 1963, Richard Held and Alan Hein tested the development of visually guided behavior in kittens reared in the dark who were placed in pairs in an illuminated carousel. One of the kittens was “passive” and could not self-locomote. The other was “active” and was free to move by itself while pulling the passive kitten through a transmission mechanism that produced an equivalent visual stimulation. The kittens experienced this condition for 3 h a day for a period of 8 weeks. They were then tested on the visual cliff. Unlike active kittens, passive kittens did not show evidence of differentiating the shallow edge of the cliff from the apparent drop. From this, Held and Hein concluded that self-generated movements are crucial for the development of visual perceptual skills. The experiment is often mentioned in the enactivist and ecological psychology literature (e.g., Gibson, 1969 ; Varela et al., 1991 ; Noë, 2004 ; Heras-Escribano, 2019 ).

This was not an isolated study. It was part of an extensive research program led by Richard Marx Held (1922–2016) during the 1950s and 1960s. We think that the hypotheses, the innovative experimental designs, and the discussions provoked by these studies are still relevant today. Through a brief review of Held’s work, we seek to contribute to discussions about the role of self-generated movements in perceptual learning. Is the self-generation of movements, or “active movements,” a necessary condition for acquiring new perceptual abilities? Or is it possible to learn to perceive without moving actively, i.e., or through “passive,” externally controlled movements?

Our main conclusion will be that many of the terms in these seemingly clear questions require clarification. In particular, we qualify the terms “active” and “passive” because the discussion will lead us to examine how these notions have been used in experimental psychology and neuroscience. We suggest that what is typically taken as a binary distinction is in fact a spectrum of possibilities with various degrees and different dimensions of activity and passivity. This may be one reason why it is sometimes difficult to replicate experimental results and reach widely accepted conclusions regarding the role of action in perception. We will draw on recent work on the sense of agency to refine the active-passive distinction. We will also offer three distinct meanings for claims concerning the role of activity in perceptual learning.

Held’s Studies on Sensory Adaptation

Participants in sensory adaptation experiments use interfaces that induce perceptual changes. For example, combinations of special lenses, prisms, and mirrors. The sensory disruption puts in evidence relations between action and perception that otherwise remain hidden. In order to learn to behave and perceive correctly again (in the cases when adaptation occurs), participants must modify their repertoire of sensorimotor schemes. Hermann von Helmholtz and George Stratton performed pioneering studies in visual sensory adaptation in the nineteenth century ( Welch, 1974 ). A few decades later Ewert (1930) ; Peterson and Peterson (1938) carried out further studies involving long-term exposure and adaptation. The period ranging from the late 1950s to the early 1970s was particularly fruitful with researchers like John G. Taylor, Irving Rock, Hans Wallach making important contributions ( Welch, 1974 , 1978 ). According to Welch, the increase in interest may be traced to two sources: the publication by Kohler (1964) of the extensive investigations that he and Theodor Erismann had carried out at the University of Innsbruck, and the work by Held and collaborators. At that time, the team led by Held was setting the pace of the investigation. This period covers his time at Brandeis University (1953–1962) and his first stage at MIT (1962–1971).

During the period in question Held was devoted to the study of adaptation to spatially inverted, reversed, and displaced perception. In most cases he used a technique called rearrangement , which consisted in presenting participants with a deliberate distortion of visual or auditory signals through special sensory interfaces. Contemporary studies in sensory adaptation continue to use variations of this method (e.g., Pfordresher and Kulpa, 2011 ; Bermejo et al., 2020 ). In typical experiments, Held exposed participants to rearrangement in pretest-posttest designs. To evaluate the effect of self-movement, participants generally underwent an active condition, in which they could move by themselves, and a passive condition, where an experimenter produced in the participant movements equivalent to the active condition. Results, replicated over a series of studies, showed that participants almost invariably compensated for the errors induced by rearrangement only or much more reliably in the active condition (e.g., Held and Hein, 1958 ; Held and Bossom, 1961 ; Held and Rekosh, 1963 ; Held and Mikaelian, 1964 ).

It is worth taking a brief look at some of these experiments. In one of his first studies, Held (1955) evaluated the effect of experiencing spatially distorted sound cues on sound localization. He used a “pseudophone” that modified the sound streams arriving at the ears causing perceived sounds to be displaced to the left or right. Participants had to orient toward a sound source before and after having practiced with the pseudophone in conditions of self-locomotion. Results showed shifts of auditory localization responses that evidenced a correction of the error induced by the device. Held suggested that adaptation happened because participants were able to associate new interaural patterns with their own movements toward the source. This early formulation of his hypothesis led him to include a passive group in subsequent designs.

In an experiment looking at visual rearrangement, Held and Hein (1958) asked participants to mark with a pencil the apparent location of the corners of a square. They could see their own hands through a prism that produced a lateral visual displacement under different conditions: self-produced hand movement, passive hand movement guided by an experimenter, and no hand movement. The only condition in which the participants were able to compensate for visual displacement was when they could move actively. Held and Bossom (1961) extended the prism rearrangement situation to locomotion. They found that an equivalent correction effect in visual direction-finding occurred when participants performed active self-locomotion. They did not compensate if they were transported by an experimenter in a wheelchair. In a follow-up experiment, Held and Mikaelian (1964) evaluated whether the lack of adaptation in the passive condition could be attributed only to the passivity involved in not being able to initiate movements or in the lack of specificity between movement and contingent stimulation. They replicated Held and Bossom’s (1961) study with the difference that they allowed participants to control the movement of the wheelchair. Adaptation occurred when participants walked by themselves but not when they used the wheelchair. The authors concluded that motor-sensory feedback must correspond to the specific behavior undergoing adaptation, i.e., walking for visual egocentric localization.

These and other studies pointed to similar conclusions 1 , which Held (1965) explained in terms of the active regulation of the plasticity of sensorimotor systems. Situations of rearrangement degrade established patterns of sensorimotor coordination. Under typical conditions, the relation between self-produced movements and stable parts of the environment is univocal and lawful. Each movement has its unique train of sensory stimulations and the perceiver is adapted to this lawful relation. A rearrangement situation alters this relation, confounding participants. A given familiar movement will have unexpected sensory consequences, and expected sensory consequences can only be obtained by the performance of unfamiliar movements. New lawful sensorimotor relations may exist in the rearranged situation but these are not yet obvious to the perceiver. This leads to an increase in performance variability and a decrease in accuracy. As the perceiver explores and practices movements in the rearranged situation, progressive shifts in coordination compensate for the errors induced by the atypical conditions. Gradually, as the new sensorimotor regularities are learned, performance becomes more robust and accurate. This same logic also explains the presence of aftereffects: returning to the original condition after prolonged exposure to the rearranged situation gives rise to errors similar, but in the opposite direction, to those initially induced by rearrangement.

Held consistently finds that self-produced movements are key to adaptation. When movements are passive, the condition of a degraded lawful relation between action and perception does not manifest in the same manner, and consequently adaptation does not occur reliably ( Held and Freedman, 1963 ). “Only an organism that can take account of the output signals to its own musculature is in a position to detect and factor out the rearrangement effects of both moving objects and externally imposed body movement” ( Held, 1965 , p. 92).

In this context, a key concept is that of re-afferent signals, i.e., stimulation caused by self-produced movements. Exafference, in contrast, is the stimulation that is not induced by the agent’s own movements. Both of these terms are taken from the “efference copy” formulation by von Holst and Mittelstaedt (1950) , an influential model explaining the possible mechanisms involved in perceiving the stable attributes of space.

Another key concept is correlation, which is used to describe the relationship between self-produced movements and re-afference. Because in a stable environment there is a unique feedback signal for any particular movement, it is expected that as time passes, the cumulatively experienced efferent and re-afferent signals will show high correlation. The quasi-constant relationships in correlations give rise to the idea of a motor-sensory feedback loop. Sensorimotor coordination of perceptual systems is grounded in the information entailed in these loops ( Held and Freedman, 1963 ).

Held and Freedman (1963) considered these adaptive mechanisms to underlie perceptual learning at any stage of life. At this point, the proposal became a “general theory of the plastic sensorimotor systems” ( Held and Freedman, 1963 , p. 455) in which the same sensorimotor coordination mechanism is involved in three processes: “(1) the development of normal sensory-motor control in the young, (2) the maintenance of that control once it has developed and (3) the adaptation to changes or apparent changes in the data reported by the senses of sight and hearing” ( Held, 1965 , p. 84).

Held and colleagues tested this theory by looking at sensorimotor development in young animals. The already mentioned kittens study by Held and Hein (1963) was one in this experimental series. In follow-up studies, Hein et al. (1970) evaluated three hypotheses derived from the carousel experiment. First, they investigated whether the deficit of the passive kittens in the acquisition of visually controlled behavior was due to a facilitating effect of self-locomotion or whether it was impeded by passive transport. To test this, they implemented a similar set-up, with the difference that after being exposed to the passive condition, kittens were allowed to experience the active role for a few hours a day over several days. The results showed that previous exposure to passive transport led to a significant delay in the number of hours needed to acquire a simple visuo-motor capacity (limb extension towards an approaching broad surface) with respect to the group that could always self-locomote.

The authors conducted a second study to determine if passive kittens failed in the visual cliff due to a generalized inhibition of locomotion behavior in response to visual stimuli. In this case, they exposed each eye under a different condition: “periods when visual stimulation of one eye accompanied self-produced movement alternated with periods when visual stimulation was provided to the other eye during passive transport” ( Hein et al., 1970 , p. 184). As predicted, when kittens could use the eye exposed in the active condition, they were successful both in extension response tests and in the visual cliff. When the same kittens had to use the other eye, they failed in both tasks.

A third experiment extended a study by Hein and Held (1967) . When kittens who are prevented from seeing their paws during rearing are carried down toward the edge of a horizontal surface, they show fragmented visual control of their forelimbs (extension but not accurate guidance). Failure to develop guided reaching did not affect the use of the limbs for visually guided locomotion, suggesting that reaching is a separate kind of visuo-motor coordination. In the new experiment, Hein et al. (1970) tested whether visually guided reaching might be acquired independently by each eye. The results showed that when the kittens used the eye that had not received visual input from their own limbs, they failed in visually guided reaching and succeeded when using the other eye. Visually guided reaching did not transfer interocularly.

These follow-up studies support and refine Held’s proposal concerning the enabling role of active movements in perceptual learning, showing that the control of visually guided behavior can be acquired independently for each eye and that an unsystematic correlation between self-movement and visual stimulation produces disruptive effects.

Prior to the kittens experiments, Held made an attempt to adapt his theoretical perspective to von Holst and Mittelstaedt’s (1950) “efference copy” model. Briefly, this model proposes that at the time of motor preparation, copies of the efferent motor information are used to calculate the sensory changes expected as a consequence of its execution. After the effective execution, a real proprioceptive feedback is generated, which will balance the predicted sensory feedback at the level of a so-called “Comparator.” In the case of not finding differences between the two signals, the changes in the afferences would be due to the re-afferences. Otherwise, the changes would be the result of the exafferences. Held (1961) proposed to supplement this model with mechanisms that account for the possibility of adapting to rearrangement situations. To do this, he conceived of an instance before the Comparator stage, which he called Correlation Storage. This module would be responsible for retaining combinations of concurrent efferent and re-afferent signals. The selection of a determinate efferent signal activates a combination that then passes on to the Comparator. In cases of adaptation, correlations of signals must be permanently updated. If the combinations are systematically changed it may happen that the same efferent signal can activate different combinations. This ambiguity will be gradually eliminated as more recent combinations gain more weight.

Evidently this model distances Held’s proposal from perspectives such as ecological psychology and enactivism due to its reliance on internalist explanations. However, Held himself did not make much explicit use of this model. To understand his proposal more fully, it is necessary to take into account that:

• (a) In the 1950s and 1960s models such as the “efference copy” were in full swing. Simultaneously, von Holst and Mittelstaedt (1950) and Sperry (1950) had proposed equivalent models and researchers such as Hans-Lukas Teuber, who was head of Held’s department at MIT, were working on these theories. It is likely that this context encouraged Held to reformulate his ideas in terms of comparator models.
• (b) The model was only described in his first theoretical work ( Held, 1961 ). In future articles, he did not return to it. Since then his explanations were formulated in terms of self-produced movements, reafference, and the correlation between them, without any mention of instances such as Correlation Storage.
• (c) Held clearly rejected internalist interpretations. Immediately after describing the model in the 1961 paper, he remarked that his view of perception does not fit with the idea of the information processor:

“The proposed system neither selects nor filters the incoming signals it receives on the basis of special functional relations or orderings (other than temporal) between the efferent and afferent signals. We can hope that the model forestalls assertions that the nervous system actively seeks a special kind of order which it may store for future reference. Such statements seem to me to beg the issue by assuming an internal “intelligence” to accomplish precisely what requires explanation. (…) If models of the sort proposed convincingly account for adaptive, as well as non-adaptive, psychological processes, we need have recourse neither to a mysterious internal “intelligence” that somewhat knows how to recognize and select useful sensory information nor to the equally mysterious external “intelligence” that manages to reinforce just those responses to sensory stimuli that will prove useful to the organism” ( Held, 1961 , pp. 31–32).

We may summarize Held’s proposal for explaining sensory adaptation by these claims:

• • Self-produced movements are necessary for developing a perceptual ability.
• • The stable perception of the environment relies on the consistent coordination of sensorimotor loops.
• • Sensory adaptation involves, at first, the disruption of previously established sensorimotor coordination. This is followed by the active gradual reconstruction of new stable sensorimotor patterns.
• • The structure of reconstructed sensorimotor loops is constrained by bodily and environmental features and depends on the particular patterns of active practice.
• • These principles are not restricted to the visual system, but form part of a general model of perceptual learning.

Some Reactions to Held’s Proposal

Given the interest raised by Held’s experiments, many contemporary and later studies attempted to replicate their results. Some of them had difficulty doing so. Some found no significant differences in the adaptation achieved between the conditions of passive and active movements ( Weinstein et al., 1964 ; Pick and Hay, 1965 ; Singer and Day, 1966 ; Fishkin, 1969 ; Foley and Maynes, 1969 ; Baily, 1972 ; Gyr et al., 1979 ). There were even studies that showed the possibility of adaptation without movements at all ( Howard et al., 1965 ; Kravitz and Wallach, 1966 ). The kittens study was particularly difficult to replicate. Walk et al. (1988) wondered if passive kittens paid less attention than active ones to the environment through which they were transported. They repeated the experiment with kittens that either were given something interesting to watch (toy cars racing on a track) or were allowed to move through the environment by lifting their heads to close a microswitch that operated a go-cart. Kittens that paid greater attention to the environment discriminated depth on the visual cliff, whereas those reared with similar light exposure conditions but without the increased attention did not discriminate. The results confirmed the authors’ hypothesis, passive kittens paid less attention than active kittens to the environment and probably this, and not the absence of self-generated movements, enables the learning of spatial skills. It is worth clarifying that kittens were “passive” in the sense used by Held, since they could not walk, but they were “active” in the sense that they paid attention to relevant/interesting events in the environment, being able to move their head or move according to their control. We will return to this point later.

Another point of confrontation with Held’s work concerned rearrangement experiments that involved adapting hand-eye coordination, for example, marking with a pencil the position of a cross that was seen through prisms. Several studies indicated that the perceived discrepancy between the optical image and the postural sensation of the hand was an important cue not taken into account in Held’s explanation of adaptation ( Kravitz and Wallach, 1966 ; Wallach, 1968 ; Lackner, 1974 ). According to these studies, the discrepancy is constant and independent whether the hand is moved passively or actively. Therefore, there is no need to re-correlate motor information. The problem is actually one of detecting the constant discrepancy between the true and the seen position of the limb. Adaptation consists in resolving these visual-spatial conflicts. This process involves a great range of factors and subsumes elements like attention, that affect the accuracy of registering the mentioned discrepancies ( Lackner, 1981 ).

Other action-based perspectives at that time, such as incipient work in ecological psychology, also showed a distancing from Held’s proposal. James Gibson’s early empirical work on sensory adaptation was not equivalent to Held’s. It was more focused on what he called phenomenal adaptation. He had developed diverse experiments on adaptation to negative aftereffects ( Gibson, 1933 , 1937 ; Gibson and Radner, 1937 ). These were simple effects showing that during prolonged looking, adaptation can occur in the perceived curvature and tilt of static lines, the curves tending to become straighter and the tilts tending to become either more vertical or horizontal. These perceptual adaptations, according to Gibson, were a sort of phenomenal normalization to the usual conditions of the physical environment. Beyond these experiments, Gibson’s theoretical developments are closely related to “behavioral” adaptations, as he called them. In particular, he did not think that there were important differences in the perceptual information that could be obtained with self-generated movements and passive movements. While he argued that to perceive it is necessary to establish sensorimotor invariants ( Gibson, 1979 ), it seems that the movements activating those invariants could also be passive. The concept of information flow, understood as the pattern of change sensed by the perceiver ( Gibson, 1950 ), for instance, offers the possibility of explaining sensory adaptation processes through passive movements. According to Gibson, in relation to the sensory adaptation experiments:

“Head-movements would be necessary for isolating these new invariants; perhaps voluntary head movements would help in directing attention to them but passive movements should be sufficient. […] In short, according to this formula there is a way of “finding out” about the environment without necessarily behaving in the sense of performing or executing actions.” ( Gibson, 1967/1997 , §. 9).

Gibson’s position can also be understood by his considerations on proprioception and exteroception, which he did not consider as two fundamentally different and separate channels of information. The exteroceptors, such as the retina, are sensitive both to changes in the direction of objects that move in the environment and to the flow of patterns produced by the movement of the head. Meanwhile, proprioceptors can account for both externally imposed passive movements and self-generated ones. Gibson abandons the notion of independent and purely exteroceptive or proprioceptive fields, and the main problem becomes one of exhaustively defining the entire array of stimulation, irrespective of the particular receptors involved ( Cohen, 1981 ).

“Proprioception considered as the obtaining of information about one’s own action does not necessarily depend on proprioceptors, and exteroception considered as the obtaining of information about extrinsic events does not necessarily depend on exteroceptors.” ( Gibson, 1966 , p. 34).

Attentive to Gibson’s work on phenomenal adaptation (in curved and tilted lines), Held and colleagues expanded the study of these effects by exploring the possibility they may be enabled by a motor component as well. In two similar studies, they assessed adaptation to tilt ( Mikaelian and Held, 1964 ) and curvature ( Held and Rekosh, 1963 ) under active and passive conditions using specific prisms that modify these properties. Both studies were very ingenious; participants had to estimate the state of lines before and after exploring a scene with a random array of small spots. This has the effect of removing from the visual environment any lines or curves that could provide normalizing visual cues for straightness or vertical/horizontal orientation. In the active condition, participants could self-locomote with the goggles, while in the passive one they were transported by an experimenter on a wheelchair or a cart along the same route. Both studies confirmed Held’s proposal, only self-generated movements under the transformed condition induced by the goggles make participants compensate for the errors due to the prism. On removing the goggles, they experience an aftereffect of the same magnitude, but in the opposite direction as the prismatic distortion. In the passive conditions the after-effect is much smaller. These findings imply that even processes classified as purely visual or phenomenal, such as adaptation due to normalization effects, can also involve motor factors.

Discussions at the time seem to arrive at the consensus that self-generated movements were not essential to achieve sensory adaptation although their presence could facilitate the adaptive process ( Welch, 1978 ). Irving Rock summarized the state of affairs: “[self-generated] movement is important only because it allows for certain kinds of information to be registered, not because movement per se is necessary” ( Rock, 1966 , p. 42). From the 1970s onward Richard Held abandoned the study of sensory adaptation, as he describes in his autobiography ( Held, 2008 ).

Later Repercussions of Held’s Work

After a period of relative quiet, today Held’s proposal resonates with contemporary action-based theories of perception. The idea has a strong affinity with the enactive approach according to which perception relies on the mastery of lawful sensorimotor regularities, with self-generated movements being an enabling (and possibly constitutive) condition for developing these sensorimotor schemes or contingencies ( O’Regan and Noë, 2001 ; Noë, 2004 ; Di Paolo et al., 2017 ). Varela et al. (1991) brought the kittens experiment to the attention of a younger generation: “This beautiful study supports the enactive view that objects are not seen by the visual extraction of features but rather by the visual guidance of action” (pp. 174–175). This view was inspirational for the field of evolutionary robotics during the 1990s and 2000s, where the role of self-generated activity could easily be appreciated in concrete examples ( Nolfi and Floreano, 2000 ; Harvey et al., 2005 ; Vargas et al., 2014 ) particularly in models where perceptual information is generated by a robot’s own activity as in the case of a wheeled robot that uses its own angular velocity while circling objects to discriminate their size ( Pfeifer and Scheier, 1999 ). Self-generated activity was explicitly investigated in developmental robotics, as in a replication of the kittens experiment using real mobile robots ( Suzuki et al., 2005 ).

The fit of Held’s ideas with the contemporary ecological perspective is less clear. Mention of Held’s work is cursory or absent from many important books in this tradition (e.g., Lombardo, 1987 ; Reed, 1996 ; Heft, 2001 ; Chemero, 2009 ; Turvey, 2019 ), although it enjoyed some recognition in ecological theories of development (see below). While it is still argued that in order to perceive it is necessary to establish sensorimotor invariant rules ( Lobo et al., 2018 ), it is difficult to find similar pronouncements on the explanatory status of self-generated movements.

Mossio and Taraborelli (2008) have discussed these differences between the two schools. Ecological theories conceive of sensorimotor invariants as a specific transformation of information linked to the perceiver’s motion, but their presence and structure do not rely on the voluntary execution of specific motor schemes. By contrast, for the enactive approach motor schemes are intrinsic constituents of perceptual invariants, and schemes are themselves always constituted by bodily and environmental processes. Perception involves the activation of a network of actual and virtual powers and sensitivities, which by definition cannot be entirely passive, and in the case of honing perceptual skills adaptation processes cannot be divorced from activity of some kind ( Di Paolo et al., 2017 ). Ecological psychologists focus on the transformation in the ambient array, while enactivists on the changes produced by active sensorimotor schemes. It is a simplification, but it may be helpful to understand this difference to think that for ecological psychologists the paradigmatic case from which perception in general is explained is vision, e.g., appreciating the affordances of a complex scene, while for enactivists it is touch, e.g., perceiving the softness of a sponge by squeezing it.

This crude characterization serves as a first step in differentiating the enactive and ecological positions. Experimental evidence, however, belies this simple picture. For example, parallax and depth perception in cases of ambiguous optic flows are robustly dependent on voluntary movements by the observer ( Wexler and van Boxtel, 2005 ). Wexler (2003) studied the perception of ambiguous optic flows under voluntary (self-produced head movement), involuntary (movement controlled by an experimenter), and mismatch displacement conditions (the participant moves a wheelchair with her hands). These conditions help disentangle motor signals for action initiation (assumed to be available only in voluntary motion) and proprioceptive and vestibular information. The same optic flow information leads to different perceptions in voluntary vs. mismatch and vs. passive conditions. Wexler observes that the difference cannot be due to external flows or to proprioception alone but depends on the motor command. However, “it is not the mere presence of a corollary discharge, but the details of the motor command that are crucial to spatial vision” (p. 344). Wexler concludes that not only can we not disregard self-motion in spatial vision, but “the observer’s active role in initiating and producing that motion is [also] crucial” (p. 344). This evidence, that could be taken to support an enactive interpretation, could also be interpreted ecologically if the sources of integrated information are extended to include motor commands and other somatosensory signals.

Studies like these show that it is difficult to attribute empirical results as supporting either the simplified versions of the enactive or ecological positions. The evidence used often arises from perceptual situations where sensorimotor schemes are already consolidated. One way to force a contrast between the two perspectives is to analyze the development of sensorimotor invariants or contingencies. Trying to answer how sensorimotor invariants achieve a stable structure can shed light on the role of self-generated movements.

To clarify the discussion, we use the terminology introduced by De Jaegher et al. (2010) to describe different kinds of causal relations giving rise to a phenomenon. A contextual factor is one whose alteration changes the manifestation of a phenomenon, but not whether it is manifested or not, an enabling factor is causally necessary for a phenomenon to occur, and a constitutive factor is an inextricable part of what makes the phenomenon what it is. Accordingly, for ecological perspectives self-generated motion has a contextual or at most an enabling or instrumental role in the specification of invariants, while for the enactive perspective it has an enabling or even constitutive role in the consolidation of sensorimotor schemes.

From an enactive perspective, perception is constituted by the skillful use of regularities that govern the ongoing coupling between motor and sensory activity, i.e., sensorimotor contingencies ( Noë, 2004 ). Sensorimotor disruptions present the perceiver with radical obstacles and lacunae. Her established sensorimotor skills suddenly cease to make sense. Perceptual adaptation consists in the re-equilibration of sensorimotor schemes guided by engagement in a particular activity and the norms it defines, e.g., whether or not the intended object is reached successfully, quickly, efficiently, and so on. While the changes involved are more radical, the processes themselves are, as in Held’s proposal, continuous with those of ongoing equilibration involved in minor adjustments and recalibration ( Di Paolo et al., 2014 , 2017 ). A key element of adaptation and perceptual learning is the need for self-generated activity by the agent. New sensorimotor schemes cannot be learned unless the agent engages the world actively and confronts various breakdowns and tries to recover from them, all within the normative context of the activity itself. The degree and speed of adaptation will depend on whether the new normative situation is radically novel or a modification of established habits. This may explain why adaptation is not observed in conditions of mismatched voluntary movement in experiments by Held and Mikaelian (1964) where participants able to walk normally are asked to move themselves in a wheelchair. The time for adapting to a task with a very different normativity (stipulating the appropriate energy, reach, duration, and coordination of movements) can be expected to be significantly longer. Other things being equal, the enactive approach predicts that adaptation will eventually occur even in such cases (and that people with experience controlling wheelchairs will adapt faster).

Sensory adaptation experiments, either following the rearrangement paradigm or in cases of sensory substitution, are frequently cited in support of the enactive account ( Varela et al., 1991 ; Myin and Degenaar, 2014 ). Situations of rearrangement make apparent the established relation between sensorimotor schemes and perception by distorting it. In order to perceive correctly again this relation must be re-established. The perceiver must modify her sensorimotor schemes. Action and perception are constituted as an internal relation between two terms: the perceiver’s repertoire of skills and sensitivities, and the environment. (By internal relation we mean a relation on which the related terms depend and through which they change together, as opposed to an external relation between already defined and fixed entities.) This relation is enacted in concrete form in each particular situation involving current and historical environmental constraints and opportunities, goals, motivations, and so on. Changing either term induces a breakdown in the organization of this relation. This breakdown cannot be recovered exclusively from the agent’s side or from the environment’s side because a new meaningful coherence must be found between two terms in dynamic flux. For this reason, perceptual adaptation is neither the construction of new correlations that internally rearrange environmental data nor the discovery of pregiven invariants in the environment. Rather, it is the result of re-organizing the internal relation between these terms by simultaneously engaging agents and the world. It is eminently a practical, rather than an intellectual, process.

Perceptual learning demands active engagement by the agent with the environment on two related counts: (1) as the process through which new regularities can be explored and equilibrated, and (2) for establishing situated norms that regulate equilibration (otherwise there is no way for the agent to know what counts as success or failure and no way to produce new coherent, task-dependent, sensorimotor relations). For each of these reasons, one causal, the other specifying of what to learn, self-generated activity is both enabling and constitutive of perceptual learning. This activity will in general take the form of combined voluntary overt motor actions such as locomotion, and covert activity such as modulating the focus of attention. It may even take place largely through non-overt activity, as in the case of patients with locked-in syndrome learning to use a brain-computer interface ( Kyselo and Di Paolo, 2015 ).

Ecological psychology studies on recalibration have discussed the role of activity in terms of its relation with environmental flow. For example, Rieser et al. (1995) induced discrepancies between a participant’s walking speed on a treadmill and the rate of optic flow as the treadmill is dragged by a tractor. After these experiences, they showed evidence of perceptual recalibration that cannot be explained by considering motor activity or environmental flow separately. Instead, it is the product of the participant’s sensitivity to the covariance between the two. The authors state that it is difficult to disentangle the precise influence of each variable and that “much work remains to specify the biomechanical information. For example, is efference important?…Is reafference important…?” (p. 496).

Self-generated activity has been more explicitly recognized in ecological approaches to development. Eleanor Gibson seems to have been more sympathetic to the implications of Held’s work than her husband for whom, as we have seen, self-generated movements could be helpful in perception, but not necessary as such. Eleanor Gibson’s approach to perceptual learning and development is indeed compatible with James Gibson’s ideas, but puts more emphasis on the importance of the organism’s active role in exploring the environment ( Adolph and Kretch, 2015 ). Animal and environment are considered as an interactive reciprocal system in which self-produced movement provides dynamic simultaneous information about oneself and environmental events ( Gibson and Pick, 2000 ; Szokolszky et al., 2019 ). Eleanor Gibson considered locomotion as one of the major organizing behavioral systems in infancy, which allows the learning of many affordances. According to her view, perceptual development implies learning to detect new affordances as action capabilities change due to changes in the body ( Gibson, 1992 ). In a mutual relation that unfolds developmentally, efficient visually controlled locomotion involves perceiving what a given surface affords, and detecting the information that specifies this affordance requires experience in guiding the body. This experience plays a critical role in perceiving the affordance of a surface for locomotion ( Gibson and Pick, 2000 ). Eleanor Gibson was overall more receptive to discussing and accepting the implications of Held’s work (e.g., Gibson, 1969 ). She mentions the kittens study in support of her own views:

“Self-produced movement, while guiding locomotion visually, emerged as a critical factor in research with kittens by Held and Hein (1963) […] This finding strengthens the notion that guided action combining visual and kinesthetic information from the action systems involved is essential for the kind of affordance that is being learned” ( Gibson and Pick, 2000 , p. 113).

Along similar lines, Karen Adoph’s studies on infant locomotion led her to the view that a period of self-produced experience is needed to learn to perceive affordances and avoid the visual cliff ( Adolph, 2000 ; Kretch and Adolph, 2013 ). The learning experience gained with attaining a given posture (e.g., avoiding risky staircases when crawling) is not automatically transferred when a new motor ability is acquired (e.g., walking). It is necessary to learn to perceive the affordances involved in each case, since the perceptual consequences of moving while crawling or walking are very different. To say that self-produced activity is needed, crucial, or essential for perceptual development amounts to assigning it an explanatory role that is not merely contextual , but also enabling , i.e., without this activity, perceptual learning would not occur.

More recently, Jacobs and Michaels (2007) have proposed a direct learning approach whereby adaptive changes in perception occur without mediating inferential processing. The theory formally describes the information for learning as a vector field covering the space of all the perception-action couplings that can be used to perform an action. Each point in this space corresponds to a specific coupling. Changes during learning are represented as paths. Perceptual learning in this model involves three processes: the education of intention, the education of attention, and calibration. Through the education of intention agents in a situation with multiple alternatives can improve “in choosing which of the possible perceptions and actions they intend to actualize” (p. 326). The education of attention, a term taken from Gibson (1979) , is the process of learning to detect the most useful informational variable, even if intention does not change. Calibration consists of changes in the way the informational variable that is operative at a particular moment is used in perception or action. This model is meant to explain a wide range of phenomena in ecologically relevant and informationally rich situations as well as in simpler experimental situations. Although it does not explicitly address the issue of self-generated movements, it makes a clear reference to the active role of the agent. A recent reading of this work suggests that there is an equivalence between the model and the enactive proposal described by Di Paolo et al. (2017) . The similarities include, among others, the point that direct learning requires the active role of the perceiver through the perception-action coupling ( Higueras-Herbada et al., 2019 ).

Why do we find different positions within the ecological perspective in relation to Held’s proposal? We believe that this is because theories of perceptual learning ( Gibson, 1969 ; Gibson and Pick, 2000 ; Jacobs and Michaels, 2007 ) make use of a concept of agency that is not emphasized in more orthodox ecological positions (beyond the recognition that the determinants of behavior are not exclusively environmental, see Withagen et al., 2012 ). This concept has emerged more explicitly in recent decades and is part of ongoing discussions within ecological psychology. In a keynote address on the future of psychology published in 1994, Eleanor Gibson referred to agency as one of the hallmarks of human behavior. She uses the term to describe the case when an organism manifests at least some autonomy and control ( Gibson, 1994 ). Agency, according to her, is manifested in human behavior together with three other fundamental hallmarks: prospectivity, retrospectivity, and flexibility. Prospectivity and retrospectivity help define a particular animal’s region of controllable agency. Prospectivity directs action and attention toward the emerging features of situations. Retrospectivity enables agents to coordinate past experiences with present control. Flexibility in action control refers to the interchangeability of means to achieve the ends of action. From these elements Edward Reed (1996) points out that the actions of agents are not the effects of just any previous cause. “Their actions are part of a stream of regulatory activities that are typically self-initiated and modified and regulated by both internal and external factors.” ( Reed, 1996 , p. 19). Such ideas are consistent with Chemero’s (2009) proposal that we should not think of affordances as dispositional properties. We should understand them in relational terms instead. Chemero believes that perception and action should always be considered in the context of the agent-environment system. To understand the relationship that an agent establishes with her environment, it is not enough to simply focus on the constraints and regularities that may exist, it is also necessary to focus on how the agent is able to selectively be sensitive to or be invited by some affordances and not others ( Bruineberg et al., 2019 ). In making agency an important concept as well as a topic for further research, this ecological strand finds much in common with the enactive approach, for which the idea of agency is central (e.g., Di Paolo, 2005 ; Barandiaran et al., 2009 ; Di Paolo et al., 2017 ).

The discussion today, as it was in the 1960s and thereafter, is fraught with difficulties that arise from the use of apparently straightforward formulations in the context of very complex phenomena. Held proposed that self-generated movements are necessary for achieving sensory adaptation, for perceptual learning and, in general terms, for perceiving in a stable manner. To this we have seen a range of responses that go from the flat denial that self-generated movements (or movement at all) play a role in perceptual learning, to James Gibson’s interpretation that they may facilitate attention but are not really necessary, to Eleanor Gibson and colleagues suggesting they are indeed necessary as part of a mutual developmental influence between action and perception, and to the enactive view, for which an agent’s activity is not only necessary but is itself an inseparable part of the processes of perceptual learning. We have interpreted these different views in terms of contextual, enabling, and constitutive relations. Table 1 summarizes the possible positions on the causal status of self-generated movements in perceptual learning.

Summary of the different responses to Held’s proposal concerning the causal status of self-generated movements (SGM) for sensory adaptation and perceptual learning.

Contextual	SGM may facilitate perceptual learning, but learning can occur without them	J. J. Gibson, Rock, Welch, Lackner
Enabling	SGM are necessary for perceptual learning and development through reciprocal loops between action and perception. Perceptual learning is not possible without them	E. J. Gibson, Adolph, Noë, O’Regan
Constitutive	SGM, or self-generated activity in general, are an integral part of the processes of equilibration, stability, and formation of new schemes that define perceptual learning	Noë, Di Paolo et al.

We will not attempt to settle the debate here, in part because we must still critically examine the notions of activity and passivity that have been used above. As with other ideas in these discussions, this distinction is anything but simple.

Is There Ever a Purely Passive or Purely Active Condition?

Several attempts to replicate Held’s rearrangement studies were unsuccessful. There is an intrinsic difficulty in determining what counts as active and passive conditions in experimental situations. Simple operational definitions can be deceiving. It may be possible to restrict some body movements (e.g., locomotion, movements of the arm) but minor movements (e.g., head or eye movements) are more difficult to control. For instance, in the Held and Hein (1963) experiment all kittens could move their limbs, the difference was that for the passive group there was no correspondence between limb movement and displacement. Active processes that potentially influence perceptual learning can occur in situations of passivity. Indeed, several of Held’s studies show a marked individual variability in the passive condition. Although most of the passive subjects failed to adapt, some did so partially or even fully (e.g., see Held and Hein, 1958 ; Held and Bossom, 1961 ; Hein et al., 1970 ).

Even in cases where participants are completely immobile or are moved by an external force, it is very hard to account for what they are attending to. In the replication of the kittens study by Walk et al. (1988) , animals in the passive condition that were watching the toy cars could not locomote but could freely move their head. The authors downplay the role of these movements and attribute spatial learning to the visual scene that captures the kittens’ attention. These movements, however, were very frequent and enabled kittens to discover “a world in depth” ( Walk et al., 1988 , p. 251). So, even if they could not perform active locomotion, these subtle exploratory actions elicited by the kittens’ interest in the experimental situation could have contributed to learning. It can be hard to determine whether it was one or the other, head and eye movements or visual attention, that gave rise to learning. It may even be a confounded effect between these factors as presumably attention would have faded quickly if the animals could not explore the scene with head and eye movements.

Similar ambiguities arise when defining what counts as an active condition. There can be important differences in the repertoires of actions that participants can perform, from very rich to extremely poor patterns, from attentive and energetic to distracted and lethargic attitudes. As with attention, it is not always easy to ascertain the level of motivation or fatigue with which participants actively perform a task.

An anecdote from an ongoing experiment by two of the authors serves as an example for the point in question. We have performed a study to compare the effect of active and passive exploration on a task where participants, blindfolded and seated, had to reach toward a sound source located in front of them at different distances (similar task as in Hüg et al., 2019 ). In the active condition, during a training session participants freely explored the arrangement until reaching and touching the sound source. In the passive condition their arm was moved with a sling by the experimenter until the hand made contact with the source. In the posttest the “active” group showed a more precise performance. The “passive” group showed great variability; some participants improved their performance, others did not and others exhibited strange response patterns. At the end of the experiment, passive participants reported very different experiences. For example, one said that she was practically asleep. Another commented that he was extremely attentive to how his arm was being moved. In general, we could not determine these differences from mere observation.

These ambiguities are also manifested in neuroimaging studies. Passive conditions can differ significantly depending on the protocol, motivation, or attention. If passive movements are mechanically administered by a robot, the brain regions that become activated differ from those involved when an experimenter moves the body of the participant. Van de Winckel et al. (2013) suggest that this occurs because the movements performed by the experimenter are never exactly the same, which stimulates in the participant an awareness and sensory monitoring of the moved body. It is not clear if self-generated and passive movements involve the same brain regions. Some studies show that both common and different areas are activated ( Sahyoun et al., 2004 ; Ciccarelli et al., 2005 ; Van de Winckel et al., 2013 ). Others do not find any significant difference ( Weiller et al., 1996 ; Guzzetta et al., 2007 ).

What may look like a reasonable experimental operationalization can fail to capture relevant aspects of a participant’s activity. Activity does not fully stop simply because participants are instructed not to move by themselves. There is, to an extent, always an active element even in the most passive of conditions provided the participant is indeed awake and capable of regulating attention, emotion, effort, inner speech, etc. Participants in typical passive conditions accept an external control source for their movements. But this can involve active elements such as inhibiting a habitual resistance to such external interventions and remaining vigilant that movements do not become too uncomfortable. There is, in contrast, an inherently passive element in every experimental situation, no matter how freely participants may move, in that they accept and comply with the instructions they are given and do not intervene by altering the experimental set-up.

To confound matters further, attributing responsibility for action can be difficult due to social factors, not only in situations of explicit social interaction ( De Jaegher et al., 2010 ), but in general as experimental instructions, clarifications, unintended suggestions, attitudes toward experimenters, social norms, differences in culture and personality, and so on, all form part of a joint participatory construction of sense ( De Jaegher and Di Paolo, 2007 ). Allowing the intervention of another person over the initiation and regulation of our sensorimotor schemes, as in most experimental passive conditions, is a form of interaction that demands an active kind of acceptance and monitoring (see Di Paolo et al., 2018 , pp. 148–149). What a participant does or does not do in active or passive conditions is shaped by the social, linguistic, cultural, material, and technical factors at play in the experiment.

Activity and passivity should not in general be understood as forming a binary distinction. This is the case even in apparently well-defined scenarios where the distinction is applied to a restricted domain, such as whether a movement is self-generated or not. With this, we do not intend to imply that the distinction is useless and should be abandoned. Rather, we think it should be refined. We believe there are different degrees and different dimensions of activity and passivity and that articulating these differences has important theoretical and experimental implications.

A first correction that can bring some clarification is to acknowledge that the distinction is in general only a relative one. Given a series of constraints (instructions, set-up, protocol) it may be perfectly valid to describe a condition as active if it allows a significantly higher degree of choice, control, and engagement by the participant than the passive condition, and vice versa (i.e., “more/less active than …”). If constraints remain fixed, the relative difference between the conditions is expected to be maintained. However, since activity and passivity are defined relative to each other, this makes comparison between different experiments risky because of the difficulty of comparing in detail what counts as active or passive in different labs, set-ups, etc.

A more principled approach to refining the active-passive distinction results from considerations concerning the sense of agency, i.e., the aspects of lived experience that continually tell us whether we are the agents of our actions (see e.g., Synofzik et al., 2008 ; Gallagher, 2012 ). The sense of agency is illuminating not only because it involves the experienced aspects of the active-passive spectrum but also because the conceptual complexity is similar. The sense of agency is not an either-or aspect of experience, contrary to what we may think by contrasting clear-cut cases such as moving an arm or having somebody else move it for us. It is a sense with many facets not always easy to disentangle. Some aspects of the sense of agency, particularly the feeling of being involved in an action, are pre-reflective and phenomenologically recessive, that is, in normal circumstances primary awareness is with the action not with who is performing it. In cases of breakdown, interruptions, etc., however, we become more presently aware that it was the action that we ourselves have been performing (“something stopped me in my tracks”). Other aspects of the sense of agency, such as the judgment of agency can be reflective, i.e., when we take an introspective stance in the planning of an action or when monitoring its performance. These aspects can take the form of retrospective conceptual attributions (“I did that”) or ongoing deliberate regulations (“Now I must move this cursor just a bit more to the right”).

Similar differences apply to the active-passive distinction. We can expect both pre-reflective and reflective aspects to be in place, as well as differences to do with the prospective/retroactive and ongoing aspects of the action being performed. Buhrmann and Di Paolo (2017) propose a map of these differences (further elaborated in Di Paolo et al., 2017 ) and connect the phenomenological aspects with microgenetic processes involving the selection, initiation, control, and equilibration of sensorimotor schemes. The same distinctions can be applied to elucidate the active-passive distinction.

One dimension concerns action initiation. This can involve prospective intentional aspects such as an anticipatory awareness of being in a flow of activity and that a particular action needs to be executed next. At the sensorimotor level action initiation correlates with impulses to start an action as well as a sense of urge or preparation. The dimension of action initiation is to be contrasted with ongoing monitoring and control, where the relevant sense is one of progressing toward the achievement of a goal, adapting to deviations or compensating for unforeseen events and obstacles. Imaging studies confirm that different functional brain regions activate during preparation (before active movement), anticipation (prior to passive movement guided by the experimenter), and execution of movement ( Sahyoun et al., 2004 ).

In an experimental situation, it may be relatively easy to control for action initiation in distinguishing between active and passive conditions although some processes, such as preparatory neural motor potentials corresponding to the intention to move, may be active even if the ensuing movement is passive. These processes can make a difference in perception, e.g., preparation to act has been shown to affect visual discrimination ( Craighero et al., 1999 ; Fagioli et al., 2007 ). It may be less easy to establish a clear-cut difference between activity and passivity in the dimension of monitoring and control. Movement control can be effectively “handed over” to an external agent but this can be an unstable situation precisely because it is unusual and may demand actively trying not to resist the imposed movement, or trying to accompany it, or attempting to predict what the next stage is going to be. In the case of repetitive movements, the participant may fall into a regular pattern where it is unclear who is in control of the movement. Or indeed, in a fully compliant manner, the participant could be doing none of these things.

Controlling experimentally for monitoring is probably not entirely possible (adding distractions or cognitive loads may help if the point is to minimize attention but depending on the hypothesis being tested this may not be desirable). Participants’ monitoring in a passive condition may range from close scrutiny of what is going on to a total lack of attention. Again, if the task is repetitive, participants’ monitoring in the active condition may recede almost entirely as movements become automatic.

These dimensions of the active-passive distinction−initiation, control, monitoring, ( Table 2 )—and perhaps others too, should be explicitly considered in terms of experimental design and for explicating which aspects of self-generated activity are theoretically most relevant.

Dimensions of the active-passive distinction discussed in the text (there may be more).

Action/movement initiation	Prospective intentions, urge to act, reflective or pre-reflective awareness about what to do next. Preparatory motor potentials, attention to new goals.
Action/movement control	Pre-reflective sense of smooth control or obstacles and breakdowns. Adaptive equilibration via existing or newly learned schemes. Regulation via spinal circuits, etc.
Action/movement monitoring	Attention, whether focused or peripheral, to actions being performed. Ongoing (pre-reflective or reflective) verification of adequacy to intended goals.

The studies on sensory adaptation carried out by Richard Held and collaborators in the 1950s and 1960s provide us with a rich material for querying the relations between action and perception and, in particular, the role of self-generated activity in perceptual learning. The focus of heated debates at the time, this work is much less discussed today but clearly still very relevant for modern enactive and ecological psychology perspectives, and for clarifying their convergences and differences.

Many of the questions investigated by Held remain unanswered. This is partly due to the connected difficulties in clarifying the meaning of his proposal and in its empirical testing. We have introduced two refinements that throw light on the situation: one for elucidating the kinds of causal relations that may be at play, another for explicating the active-passive distinction.

To say that active, and not passive, participants demonstrate sensory adaptation in cases of rearrangement can mean different things. Self-generated activity may facilitate learning without it being strictly necessary, or it may be required for learning to occur, or it may itself be an inextricable part of the learning process. We have discussed examples of these different interpretations and proposed that they should be, respectively, categorized into contextual, enabling, and constitutive positions ( Table 1 ).

To say that a participant is active, and not passive, can also mean different things. It generally means that they are allowed to move by themselves in contrast to being moved by others. But this difference is relative and dependent on the experimental conditions. Distinct dimensions of activity can be at play in either active or passive conditions. Active movements are externally constrained by social situations, experimental instructions, and set-ups. Self-initiated activity does not necessarily stop when participants allow themselves to be moved. We have appealed to considerations regarding the sense of agency to refine the active-passive distinction and proposed that at least the following three dimensions be differentiated: action initiation, action control, and action monitoring ( Table 2 ). These are strictly dimensions and not binaries in the sense that different aspects and different degrees of intensity can be at play in each.

These considerations can help us understand apparently contradictory empirical evidence and propose more precise hypotheses. Combining the refinements summarized in Tables 1 and ​ and2 2 yields 9 possible ways of interpreting the claim that active participants adapt better to sensory rearrangement. We do not suggest that this list is exhaustive but it may be enough to help elaborate more precise ways of articulating the convergences and differences both within and between the enactive and the ecological positions.

Clarifying the active-passive distinction goes beyond the study of perception. It has implications, for instance, in areas such as motor rehabilitation. It is well established that active movement improves the recovery of motor function, but the therapeutic role of applying movements passively is still controversial ( Lindberg et al., 2004 ; Zeng et al., 2018 ; Noble et al., 2019 ). Some evidence indicates that proprioceptive input caused by passive movements (controlled by a therapist or assisted by a robot) can contribute to improving motor function through the reorganization of the cortical areas involved in sensory integration ( Carel et al., 2000 ). Others, however, state that passive movement is insufficient and active participation by the patient is required ( Hogan et al., 2006 ). Breaking down activity into the dimensions of action initiation, control, and monitoring could help make sense of these differences and knowing which aspects of activity are therapeutically important could lead to improvements in rehabilitation practices.

We conclude by highlighting again the historical and current importance of Richard Held’s rearrangement studies. His experimental designs were original and imaginative, his theoretical interpretations very innovative for the time. Either by affinity or contrast, current action-based theories of perception owe much to Held’s work. In future work, it would be beneficial to examine other theoretical proposals on adaptation and perceptual learning (such as those by Harris, 1965 ; Rock, 1966 ; Wallach, 1987 ), and the role of memory in such processes (e.g., Glenberg, 1997 ), as well as related work in computational neuroscience in the light of the classifications introduced here. The dialog between enactivists and ecological psychologists, we believe, can only benefit from the common ground that Held’s studies provide and from understanding his ideas more thoroughly.

Author Contributions

All authors conceived, designed, and wrote the sections of the manuscript and contributed to manuscript revision, read, and approved the submitted version.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

1 According to the SCOPUS database, Held with his team published 34 articles related to this topic between 1955 and 1974.

• Adolph K. E. (2000). Specificity of learning: why infants fall over a veritable cliff. Psychol. Sci. 11 290–295. 10.1111/1467-9280.00258 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Adolph K. E., Kretch K. S. (2015). Gibson’s theory of perceptual learning. Int. Encycl. Soc. Behav. Sci. 10 127–134. 10.1016/B978-0-08-097086-8.23096-1 [ CrossRef ] [ Google Scholar ]
• Baily J. S. (1972). Arm-body adaptation with passive arm movements. Percept. Psychophys. 12 39–44. 10.3758/BF03212839 [ CrossRef ] [ Google Scholar ]
• Barandiaran X. E., Di Paolo E. A., Rohde M. (2009). Defining agency: individuality, normativity, asymmetry, and spatio-temporality in action. Adapt. Behav. 17 367–386. 10.1177/1059712309343819 [ CrossRef ] [ Google Scholar ]
• Bermejo F., Di Paolo E. A., Gilberto L. G., Lunati V., Barrios M. V. (2020). Learning to find spatially reversed sounds. Sci. Rep. 10 1–14. 10.1038/s41598-020-61332-4 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bruineberg J., Chemero A., Rietveld E. (2019). General ecological information supports engagement with affordances for ‘higher’ cognition. Synthese 196 5231–5251. 10.1007/s11229-018-1716-9 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Buhrmann T., Di Paolo E. A. (2017). The sense of agency – a phenomenological consequence of enacting sensorimotor schemes. Phenomenol. Cogn. Sci. 16 207–236. 10.1007/s11097-015-9446-7 [ CrossRef ] [ Google Scholar ]
• Carel C., Loubinoux I., Boulanouar K., Manelfe C., Rascol O., Celsis P., et al. (2000). Neural substrate for the effects of passive training on sensorimotor cortical representation: a study with functional magnetic resonance imaging in healthy subjects. J. Cereb. Blood Flow Metab. 20 478–484. 10.1097/00004647-200003000-00006 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Chemero A. (2009). Radical Embodied Cognitive Science. Cambridge, MA: MIT Press. [ Google Scholar ]
• Ciccarelli O., Toosy A. T., Marsden J. F., Wheeler-Kingshott C. M., Sahyoun C., Matthews P. M., et al. (2005). Identifying brain regions for integrative sensorimotor processing with ankle movements. Exp. Brain Res. 166 31–42. 10.1007/s00221-005-2335-5 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cohen M. M. (1981). “ Visual-proprioceptive interactions ,” in Intersensory Perception and Sensory Integration , eds Walk R. D., Pick H. L., Jr. (Boston, MA: Springer; ), 175–215. 10.1007/978-1-4615-9197-9_6 [ CrossRef ] [ Google Scholar ]
• Craighero L., Fadiga L., Rizzolatti G., Umiltà C. (1999). Action for perception: a motor-visual attentional effect. J. Exp. Psychol. Hum. Percept. Perform. 25 1673–1692. 10.1037/0096-1523.25.6.1673 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• De Jaegher H., Di Paolo E. A. (2007). Participatory sense-making: an enactive approach to social cognition. Phenomenol. Cogn. Sci. 6 485–507. 10.1007/s11097-007-9076-9 [ CrossRef ] [ Google Scholar ]
• De Jaegher H., Di Paolo E. A., Gallagher S. (2010). Can social interaction constitute social cognition? Trends Cogn. Sci. 14 441–447. 10.1016/j.tics.2010.06.009 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Di Paolo E. A. (2005). Autopoiesis, adaptivity, teleology, agency. Phenomenol. Cogn. Sci. 4 429–452. 10.1007/s11097-005-9002-y [ CrossRef ] [ Google Scholar ]
• Di Paolo E. A., Barandiaran X. E., Beaton M., Buhrmann T. (2014). Learning to perceive in the sensorimotor approach: piaget’s theory of equilibration interpreted dynamically. Front. Hum. Neurosci. 8 : 551 . 10.3389/fnhum.2014.00551 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Di Paolo E. A., Buhrmann T., Barandiaran X. E. (2017). Sensorimotor Life: An Enactive Proposal. Oxford: Oxford University Press. [ Google Scholar ]
• Di Paolo E. A., Cuffari E. C., De Jaegher H. (2018). Linguistic Bodies. The Continuity Between life and Language. Cambridge, MA: MIT Press. [ Google Scholar ]
• Ewert P. H. (1930). A study of the effect of inverted retinal stimulation upon spatially coordinated behavior. Genet. Psychol. Monogr. 7 177–363. [ Google Scholar ]
• Fagioli S., Hommel B., Schubotz R. I. (2007). Intentional control of attention: action planning primes action related stimulus dimensions. Psychol. Res. 71 22–29. 10.1007/s00426-005-0033-3 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Fishkin S. M. (1969). Passive vs active exposure and other variables related to the occurrence of hand adaptation to lateral displacement. Percept. Mot. Skills 29 291–297. 10.2466/pms.1969.29.1.291 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Foley J. E., Maynes F. J. (1969). Comparison of training methods in the production of prism adaptation. J. Exp. Psychol. 81 151–155. 10.1037/h0027429 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Gallagher S. (2012). Multiple aspects in the sense of agency. New Ideas Psychol. 30 15–31. 10.1016/j.newideapsych.2010.03.003 [ CrossRef ] [ Google Scholar ]
• Gibson E. J. (1969). Principles of Perceptual Learning and Development. New York, NY: Appleton-Century-Crofts. [ Google Scholar ]
• Gibson E. J. (1992). “ How to think about perceptual learning: Twenty-five years later ,” in Cognition: Conceptual and Methodological Issues , eds Pick H. L., van den Broek P., Knill D. C. (Washington, D. C: American Psychological Association; ), 215–237. 10.1037/10564-009 [ CrossRef ] [ Google Scholar ]
• Gibson E. J. (1994). Has psychology a future? Psychol. Sci. 5 69–76. 10.1111/j.1467-9280.1994.tb00633.x [ CrossRef ] [ Google Scholar ]
• Gibson E. J., Pick A. D. (2000). Perceptual Learning and Development: An Ecological Approach to Perceptual Learning and Development. Oxford: Oxford University Press. [ Google Scholar ]
• Gibson J. J. (1933). Adaptation, after-effect and contrast in the perception of curved lines. J. Exp. Psychol. 16 1–31. 10.1037/h0074626 [ CrossRef ] [ Google Scholar ]
• Gibson J. J. (1937). Adaptation, after-effect, and contrast in the perception of tilted lines. II. Simultaneous contrast and the areal restriction of the after-effect. J. Exp. Psychol. 20 553–569. 10.1037/h0057585 [ CrossRef ] [ Google Scholar ]
• Gibson J. J. (1950). The Perception of the Visual World. Boston, MA: Houghton Mifflin, 10.2307/1419017 [ CrossRef ] [ Google Scholar ]
• Gibson J. J. (1966). The Senses Considered as Perceptual Systems. Boston, MA: Houghton Mifflin. [ Google Scholar ]
• Gibson J. J. (1979). The Ecological Approach to Visual Perception. Boston: Houghton Mifflin. [ Google Scholar ]
• Gibson J. J., Radner M. (1937). Adaptation, after-effect and contrast in the perception of tilted lines. I. Q. Stud. J. Exp. Psychol. 20 453–467. 10.1037/h0059826 [ CrossRef ] [ Google Scholar ]
• Gibson J. J. (1967/1997). “ Note on the interpretation of experiments concerned with perceptual adaptation ,” in The Purple Perils: A Selection of James Gibson’s Unpublished Essays on the Psychology of Perception , eds Pittenger J., Reed E., Kim M., Best L. ( https://commons.trincoll.edu/purpleperils/1965-1967/note-on-the-interpretation-of-experiments-concerned-with-perceptual-adaptation-for-discussion-in-perception-seminar-january-10th/ ) (accessed January 2020). [ Google Scholar ]
• Glenberg A. M. (1997). What memory is for. Behav. Brain Sci. 20 1–19. 10.1017/S0140525X97000010 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Guzzetta A., Staudt M., Petacchi E., Ehlers J., Erb M., Wilke M., et al. (2007). Brain representation of active and passive hand movements in children. Pediatr. Res. 61 485–490. 10.1203/pdr.0b013e3180332c2e [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Gyr J., Willey R., Henry A. (1979). Motor-sensory feedback and geometry of visual space: an attempted replication. Behav. Brain Sci. 2 59–64. 10.1017/S0140525X00060702 [ CrossRef ] [ Google Scholar ]
• Harris C. S. (1965). Perceptual adaptation to inverted, reversed, and displaced vision. Psychol. Rev. 72 419–444. 10.1037/h0022616 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Harvey I., Di Paolo E. A., Tuci E., Wood R., Quinn M. (2005). Evolutionary robotics: a new scientific tool for studying cognition. Artif. Life 11 79–98. 10.1162/1064546053278991 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Heft H. (2001). Ecological Psychology in Context: James Gibson, Roger Barker, and the Legacy of William James’s Radical Empiricism. Mahwah, NJ: Lawrence Erlbaum Associates. [ Google Scholar ]
• Hein A., Held R. (1967). Dissociation of the visual placing response into elicited and guided components. Science 158 390–392. 10.1126/science.158.3799.390 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hein A., Held R., Gower E. C. (1970). Development and segmentation of visually controlled movement by selective exposure during rearing. J. Comp. Physiol. Psychol. 73 181–187. 10.1037/h0030247 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Held R. (1955). Shifts in binaural localization after prolonged exposures to atypical combinations of stimuli. Am. J. Psychol. 68 526–548. 10.2307/1418782 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Held R. (1961). Exposure-history as a factor in maintaining stability of perception and coordination. J. Nervous Ment. Dis. 132 26–32. 10.1097/00005053-196101000-00005 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Held R. (1965). Plasticity in sensory-motor systems. Sci. Am. 213 89–94. 10.1038/scientificamerican1165-84 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Held R. (2008). “ Richard Held ,” in The History of Neuroscience In Autobiography , ed. Squire L. R. (New York, NY: Oxford Press; ), 158–186. [ Google Scholar ]
• Held R., Bossom J. (1961). Neonatal deprivation and adult rearrangement: complementary techniques for analyzing plastic sensory-motor coordinations. J. Comp. Physiol. Psychol. 54 33 . 10.1037/h0046207 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Held R., Freedman S. (1963). Plasticity in human sensorimotor control. Science 142 455–462. 10.1126/science.142.3591.455 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Held R., Hein A. (1958). Adaptation of disarranged hand-eye coordination contingent upon reafferent stimulation. Percept. Mot. Skills 8 87–90. 10.2466/PMS.8.3.87-90 [ CrossRef ] [ Google Scholar ]
• Held R., Hein A. (1963). Movement-produced stimulation in the development of visually guided behavior. J. Comp. Physiol. Psychol. 56 872–876. 10.1037/h0040546 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Held R., Mikaelian H. (1964). Motor-sensory feedback versus need in adaptation to rearrangement. Percept. Mot. Skills 18 685–688. 10.2466/pms.1964.18.3.685 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Held R., Rekosh J. (1963). Motor-sensory feedback and the geometry of visual space. Science 143 722–723. 10.1126/science.141.3582.722 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Heras-Escribano M. (2019). Pragmatism, enactivism, and ecological psychology: towards a unified approach to post-cognitivism. Synthese. 10.1007/s11229-019-02111-1 [ CrossRef ] [ Google Scholar ]
• Higueras-Herbada A., de Paz C., Jacobs D. M., Travieso D., Ibáñez-Gijón J. (2019). The direct learning theory: a naturalistic approach to learning for the post-cognitivist era. Adapt. Behav. 27 389–403. 10.1177/1059712319847136 [ CrossRef ] [ Google Scholar ]
• Hogan N., Krebs H. I., Rohrer B., Palazzolo J. J., Dipietro L., Fasoli S. E., et al. (2006). Motions or muscles? Some behavioral factors underlying robotic assistance of motor recovery. J. Rehabil. Res. Dev. 43 605–618. 10.1682/JRRD.2005.06.0103 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Howard I. P., Craske B., Templeton W. B. (1965). Visuomotor adaptation to discordant exafferent stimulation. J. Exp. Psychol. 70 189–191. 10.1037/h0022198 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hüg M. X., Vergara R. O., Tommasini F. C., Etchemendy P. E., Bermejo F., Fernandez L. G. (2019). Reaching measures and feedback effects in auditory peripersonal space. Scie. Rep. 9 1–14. 10.1038/s41598-019-45755-2 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jacobs D. M., Michaels C. F. (2007). Direct learning. Ecol. Psychol. 19 321–349. 10.1080/10407410701432337 [ CrossRef ] [ Google Scholar ]
• Kohler I. (1964). The Formation and Transformation of the Perceptual World. New York: International Universities Press. [ Google Scholar ]
• Kravitz J. H., Wallach H. (1966). Adaptation to displaced vision contingent upon vibrating stimulation. Psychon. Sci. 6 465–466. 10.3758/BF03328093 [ CrossRef ] [ Google Scholar ]
• Kretch K. S., Adolph K. E. (2013). Cliff or step? Posture-specific learning at the edge of a drop-off. Child Dev. 84 226–240. 10.1111/j.1467-8624.2012.01842.x [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kyselo M., Di Paolo E. A. (2015). Locked-in syndrome: a challenge for embodied cognitive science. Phenomenol. Cogn. Sci. 14 517–542. 10.1007/s11097-013-9344-9 [ CrossRef ] [ Google Scholar ]
• Lackner J. R. (1974). Adaptation to displaced vision: role of proprioception. Percept. Mot. Skills 38 1251–1256. 10.2466/pms.1974.38.3c.1251 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lackner J. R. (1981). “ Some aspects of sensory-motor control and adaptation in man ,” in Intersensory Perception and Sensory Integration , eds Walk R. D., Pick H. L. (Boston, MA: Springer; ), 143–173. 10.1007/978-1-4615-9197-9_5 [ CrossRef ] [ Google Scholar ]
• Lindberg P., Schmitz C., Forssberg H., Engardt M., Borg J. (2004). Effects of passive-active movement training on upper limb motor function and cortical activation in chronic patients with stroke: a pilot study. J. Rehabil. Med. 36 117–123. 10.1080/16501970410023434 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lobo L., Heras-Escribano M., Travieso D. (2018). The history and philosophy of ecological psychology. Front. Psychol. 9 : 2228 . 10.3389/fpsyg.2018.02228 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lombardo T. J. (1987). The Reciprocity of Perceiver and Environment: The Evolution of James J. Gibson’s Ecological Psychology. Mahwah, NJ: Lawrence Erlbaum Associates. [ Google Scholar ]
• Mikaelian H., Held R. (1964). Two types of adaptation to an optically-rotated visual field. Am. J. Psychol. 77 257–263. 10.2307/1420132 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Mossio M., Taraborelli D. (2008). Action-dependent perceptual invariants: from ecological to sensorimotor approaches. Conscious. Cogn. 17 1324–1340. 10.1016/j.concog.2007.12.003 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Myin E., Degenaar J. (2014). “ Enactive vision ,” in The Routledge Handbook of Embodied Cognition , ed. Shapiro L. A. (London: Routledge; ), 90–98. [ Google Scholar ]
• Noble S., Pearcey G. E., Quartly C., Zehr E. P. (2019). Robot controlled, continuous passive movement of the ankle reduces spinal cord excitability in participants with spasticity: a pilot study. Exp. Brain Res. 237 3207–3220. 10.1007/s00221-019-05662-4 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Noë A. (2004). Action in Perception. Cambridge: The MIT Press. [ Google Scholar ]
• Nolfi S., Floreano D. (2000). Evolutionary Robotics: Biology, Intelligence, and Technology of Self-Organizing Machines. Cambridge, MA: MIT Press. [ Google Scholar ]
• O’Regan J. K., Noë A. (2001). A sensorimotor account of vision and visual consciousness. Behav. Brain Sci. 24 939–1031. 10.1017/S0140525X01000115 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Peterson J., Peterson J. K. (1938). Does practice with inverting lenses make vision normal? Psychol. Monogr. 50 12–37. 10.1037/h0093515 [ CrossRef ] [ Google Scholar ]
• Pfeifer R., Scheier C. (1999). Understanding Intelligence. Cambridge, MA: MIT Press. [ Google Scholar ]
• Pfordresher P. Q., Kulpa J. D. (2011). The dynamics of disruption from altered auditory feedback: Further evidence for a dissociation of sequencing and timing. J. Exp. Psychol. Hum. Percept. Perform. 37 949–967. 10.1037/a0021435 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pick H. L., Jr., Hay J. C. (1965). A passive test of the Held reafference hypothesis. Percept. Mot. Skills 20 : 10702 . 10.2466/pms.1965.20.3c.1070 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pulvermüller F., Fadiga L. (2010). Active perception: sensorimotor circuits as a cortical basis for language. Nat. Rev. Neurosci. 11 351–360. 10.1038/nrn2811 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Reed E. S. (1996). Encountering the World: Toward an Ecological Psychology. New York, NY: Oxford University Press. [ Google Scholar ]
• Rieser J. J., Pick H. L., Ashmead D. H., Garing A. E. (1995). Calibration of human locomotion and models of perceptual-motor organization. J. Exp. Psychol. Hum. Percept. Perform. 21 480–497. 10.1037/0096-1523.21.3.480 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Rock I. (1966). The nature of Perceptual Adaptation. New York, NY: Basic Books. [ Google Scholar ]
• Sahyoun C., Floyer-Lea A., Johansen-Berg H., Matthews P. M. (2004). Towards an understanding of gait control: brain activation during the anticipation, preparation and execution of foot movements. Neuroimage 21 568–575. 10.1016/j.neuroimage.2003.09.065 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Singer G., Day R. H. (1966). Spatial adaptation and aftereffect with optically transformed vision: effects of active and passive responding and the relationship between test and exposure responses. J. Exp. Psychol. 71 725–731. 10.1037/h0023089 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Sperry R. W. (1950). Neural basis of the spontaneous optokinetic response produced by visual inversion. J. Comp. Physiol. Psychol. 43 482–489. 10.1037/h0055479 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Suzuki M., Floreano D., Di Paolo E. A. (2005). The contributions of active body movement to visual development in evolutionary robots. Neural Networks 18 657–666. 10.1016/j.neunet.2005.06.043 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Synofzik M., Vosgerau G., Newen A. (2008). I move, therefore I am: a new theoretical framework to investigate agency and ownership. Conscious. Cogn. 17 411–424. 10.1016/j.concog.2008.03.008 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Szokolszky A., Read C., Palatinus Z., Palatinus K. (2019). Ecological approaches to perceptual learning: learning to perceive and perceiving as learning. Adapt. Behav. 27 363–388. 10.1177/1059712319854687 [ CrossRef ] [ Google Scholar ]
• Turvey M. T. (2019). Lectures in Perception. An Ecological Perspective. London: Routledge. [ Google Scholar ]
• Van de Winckel A., Klingels K., Bruyninckx F., Wenderoth N., Peeters R., Sunaert S., et al. (2013). How does brain activation differ in children with unilateral cerebral palsy compared to typically developing children, during active and passive movements, and tactile stimulation? An fMRI study. Res. Dev. Disabil. 34 183–197. 10.1016/j.ridd.2012.07.030 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Varela F. J., Thompson E., Rosch E. (1991). The Embodied Mind: Cognitive Science and Human Experience. Cambridge, MA: MIT Press. [ Google Scholar ]
• Vargas P., Di Paolo E. A., Harvey I., Husbands P. (2014). Horizons for Evolutionary Robotics. Cambridge, MA: MIT Press. [ Google Scholar ]
• von Holst E., Mittelstaedt H. (1950). The reafference principle: interaction between the central nervous system and the periphery. Naturwissenschaften 37 464–476. 10.1007/BF00622503 [ CrossRef ] [ Google Scholar ]
• Walk R. D., Shepherd J. D., Miller D. R. (1988). Attention and the depth perception of kittens. Bull Psychon. Soc. 26 248–251. 10.3758/BF03337301 [ CrossRef ] [ Google Scholar ]
• Wallach H. (1968). “ Informational discrepancy as a basis of perceptual adaptation ,” in The Neuropsychology of Spatially Oriented Behaviour , ed. Freeman S. J. (Homewood, IL: Dorsey; ), 209–230. [ Google Scholar ]
• Wallach H. (1987). Perceiving a stable environment when one moves. Annu. Rev. Psychol. 38 1–27. 10.1146/annurev.ps.38.020187.000245 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Weiller C., Jüptner M., Fellows S., Rijntjes M., Leonhardt G., Kiebel S., et al. (1996). Brain representation of active and passive movements. Neuroimage 4 105–110. 10.1006/nimg.1996.0034 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Weinstein S., Sersen E. A., Weinstein D. S. (1964). An attempt to replicate a study of disarranged eye-hand coordination. Percept. Mot. Skills 18 629–632. 10.2466/pms.1964.18.2.629 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Welch R. B. (1974). Research on adaptation to rearranged vision: 1966–1974. Perception 3 367–392. 10.1068/p030367 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Welch R. B. (1978). Perceptual Modification: Adapting to Altered Sensory Environments. Cambridge, MA: Academic Press. [ Google Scholar ]
• Wexler M. (2003). Voluntary head movement and allocentric perception of space. Psychol. Sci. 14 340–346. 10.1111/1467-9280.14491 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Wexler M., van Boxtel J. J. A. (2005). Depth perception by the active observer. Trends Cogn. Sci. 9 431–438. 10.1016/j.tics.2005.06.018 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Withagen R., de Poel H. J., Araújo D., Pepping G.-J. (2012). Affordances can invite behavior: reconsidering the relationship between affordances and agency. New Ideas Psychol. 30 250–258. 10.1016/j.newideapsych.2011.12.003 [ CrossRef ] [ Google Scholar ]
• Zeng X., Zhu G., Zhang M., Xie S. Q. (2018). Reviewing clinical effectiveness of active training strategies of platform-based ankle rehabilitation robots. J. Healthc. Eng. 2018 : 2858294 . 10.1155/2018/2858294 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]

Edward Thorndike: The Law of Effect

Saul McLeod, PhD

Editor-in-Chief for Simply Psychology

BSc (Hons) Psychology, MRes, PhD, University of Manchester

Saul McLeod, PhD., is a qualified psychology teacher with over 18 years of experience in further and higher education. He has been published in peer-reviewed journals, including the Journal of Clinical Psychology.

Learn about our Editorial Process

Olivia Guy-Evans, MSc

Associate Editor for Simply Psychology

BSc (Hons) Psychology, MSc Psychology of Education

Olivia Guy-Evans is a writer and associate editor for Simply Psychology. She has previously worked in healthcare and educational sectors.

On This Page:

• The law of effect states that connections leading to satisfying outcomes are strengthened while those leading to unsatisfying outcomes are weakened.
• Positive emotional responses, like rewards or praise, strengthen stimulus-response connections. Unpleasant responses weaken them.
• This establishes reinforcement as central to efficient and enduring learning. Reward is more impactful than punishment.
• Connections grow most robust when appropriate associations lead to fulfilling outcomes. The “effect” generated shapes future behavioral and cognitive patterns.

Thorndike Theory

The law of effect states that behaviors followed by pleasant or rewarding consequences are more likely to be repeated, while behaviors followed by unpleasant or punishing consequences are less likely to be repeated.

The principle was introduced in the early 20th century through experiments led by Edward Thorndike, who found that positive reinforcement strengthens associations and increases the frequency of specific behaviors.

The law of effect principle developed by Edward Thorndike suggested that:

“Responses that produce a satisfying effect in a particular situation become more likely to occur again in that situation, and responses that produce a discomforting effect become less likely to occur again in that situation (Gray, 2011, p. 108–109).”

Edward Thorndike (1898) is famous in psychology for his work on learning theory that leads to the development of operant conditioning within behaviorism .

Whereas classical conditioning depends on developing associations between events, operant conditioning involves learning from the consequences of our behavior.

Skinner wasn’t the first psychologist to study learning by consequences.  Indeed, Skinner’s theory of operant conditioning is built on the ideas of Edward Thorndike.

Experimental Evidence

Thorndike studied learning in animals (usually cats).  He devised a classic experiment using a puzzle box to empirically test the laws of learning.

• Thorndike put hungry cats in cages with automatic doors that could be opened by pressing a button inside the cage. Thorndike would time how long it took the cat to escape.
• At first, when placed in the cages, the cats displayed unsystematic trial-and-error behaviors, trying to escape. They scratched, bit, and wandered around the cages without identifiable patterns.
• Thorndike would then put food outside the cages to act as a stimulus and reward.  The cats experimented with different ways to escape the puzzle box and reach the fish.
• Eventually, they would stumble upon the lever which opened the cage.  When it had escaped, the cat was put in again, and once more, the time it took to escape was noted.  In successive trials, the cats would learn that pressing the lever would have favorable consequences , and they would adopt this behavior, becoming increasingly quick at pressing the lever.
• After many repetitions of being placed in the cages (around 10-12 times), the cats learned to press the button inside their cages, which opened the doors, allowing them to escape the cage and reach the food.

Edward Thorndike put forward a  Law of Effect, which stated that any behavior that is followed by pleasant consequences is likely to be repeated, and any behavior followed by unpleasant consequences is likely to be stopped.

Critical Evaluation

Thorndike (1905) introduced the concept of reinforcement and was the first to apply psychological principles to the area of learning.

His research led to many theories and laws of learning, such as operant conditioning. Skinner (1938), like Thorndike, put animals in boxes and observed them to see what they were able to learn.

Thorndike’s theory has implications for teaching such as preparing students mentally, using drills and repetition, providing feedback and rewards, and structuring material from simple to complex.

B.F. Skinner built upon Thorndike’s principles to develop his theory of operant conditioning. Skinner’s work involved the systematic study of how the consequences of a behavior influence its frequency in the future. He introduced the concepts of reinforcement (both positive and negative) and punishment to describe how consequences can modify behavior.

The learning theories of Thorndike and Pavlov were later synthesized by Hull (1935). Thorndike’s research drove comparative psychology for fifty years, and influenced countless psychologists over that period of time, and even still today.

Criticisms 

Critiques of the theory include that it views humans too mechanistically like animals, overlooks higher reasoning, focuses too narrowly on associations, and positions the learner too passively.

Here is a summary of some of the main critiques and limitations of Thorndike’s learning theory:

• Using animals like cats and dogs in experiments is controversial when making inferences about human learning, since animal and human cognition differ.
• The theory depicts humans as mechanistic, like animals, driven by automatic trial-and-error processes. However, human learning is more complex and not entirely explained through stimulus-response connections.
• By overemphasizing associations, the theory overlooks deeper reasoning, understanding, and meaning construction involved in learning.
• Definitions and conceptual knowledge are ignored in favor of strengthening mechanistic stimulus-response bonds.
• Learners are passive receptors rather than active or creative; educators provide rigid structured curricula rather than let learners construct knowledge.
• Learners require constant external motivation and reinforcement rather than having internal drivers.
• Failures are punished, and discipline is stressed more than conceptual grasp or successful processes.
• The focus is on isolated skills, facts, and hierarchical sequencing rather than integrated understanding.
• Evaluation only measures passive responses and test performance rather than deeper learning processes or contexts.

Application of Thorndike’s Learning Theory to Students’ Learning

Thorndike’s theory, when applied to student learning, emphasizes several key factors – the role of the environment, breaking tasks into detail parts, the importance of student responses, building stimulus-response connections, utilizing prior knowledge, repetition through drills and exercises, and giving rewards/praise.

Learning is results-focused, with the measurement of observable outcomes. Errors are immediately corrected. Repetition aims to ingrain behaviors until they become habit. Rewards strengthen desired behaviors, punishment weakens undesired behaviors.

Some pitfalls in the application include teachers becoming too authoritative, one-way communication, students remaining passive, and over-reliance on rote memorization. However, his theory effectively promotes preparation, readiness, practice, feedback, praise for progress, and sequential mastery from simple to complex.

Teachers arrange hierarchical lesson materials starting from simple concepts, break down learning into parts marked by specific skill mastery, provide examples, emphasize drill/repetition activities, offer regular assessments and corrections, deliver clear brief instructions, and utilize rewards to motivate. This style is most applicable for skill acquisition requiring significant practice.

For students, the theory instills habits of repetition, progress tracking, and associate positive outcomes to effort.

It can, however, be limited if students remain passive receivers of instruction rather than active or collaborative learners. Proper application encourages student discipline while avoiding strict, punishing environments.

Additional Laws of Learning In Thorndike’s Theory 

Thorndike’s theory explains that learning is the formation of connections between stimuli and responses. The laws of learning he proposed are the law of readiness, the law of exercise, and the law of effect.

Law of Readiness

• The law of readiness states that learners must be physically and mentally prepared for learning to occur.  This includes not being hungry, sick, or having other physical distractions or discomfort.
• Mentally, learners should be inclined and motivated to acquire the new knowledge or skill. If they are uninterested or opposed to learning it, the law states they will not learn effectively.
• Learners also require certain baseline knowledge and competencies before being ready to learn advanced concepts. If those prerequisites are lacking, acquisition of new info will be difficult.
• Overall, the law emphasizes learners’ reception and orientation as key prerequisites to successful learning. The right mindset and adequate foundation enables efficient uptake of new material.

Law of Exercise

• The law of exercise states that connections are strengthened through repetition and practice. 
• Frequent trials allow errors to be corrected and neural pathways related to the knowledge/skill to become more engrained.
• As associations are reinforced through drill and rehearsal, retrieval from long term memory also becomes more efficient.
• In sum, repeated exercise of learned material cements retention and fluency over time. Forgetting happens when such connections are not actively preserved through practice.

Gray, P. (2011). Psychology (6th ed.) New York: Worth Publishers.

Hull, C. L. (1935). The conflicting psychologies of learning—a way out . Psychological Review, 42(6) , 491.

Skinner, B. F. (1938). The behavior of organisms: An experimental analysis . New York: Appleton-Century.

Thorndike, E. L. (1898). Animal intelligence: An experimental study of the associative processes in animals. Psychological Monographs: General and Applied, 2(4), i-109.

Thorndike, E. L. (1905). The elements of psychology . New York: A. G. Seiler.

Shogo Tanaka's Psychology & Philosophy Lab.

• Publications

Saturday, December 17, 2011

Classic experiment by held and hein.

blog psychology

Development of visually guided behaviour.

RICHARD HELD AND ALAN HEIN’S EXPERIMENT ON VISUALLY-GUIDED BEHAVIOUR (1963)

Introduction This is a study about kittens. Lots and lots of kittens.

did i hear… KITTENZZ!!!211″!

To understand this study, you need to learn the meaning of two terms: sensation and perception.

• Sensation is the process of bringing information from the external world into the internal body and brain.
• Perception is the process of selecting, organising and interpreting the raw information that has been taken to the brain by your senses (i.e. touch, taste, sight, sound and smell).

Held and Hein were interested in the development of perception. The nature vs nurture debate is one of the most common means used to research the topic of perception. To learn more about the influence of nature and nurture, psychologists have studied human babies, cataract patients, animals, different cultures and adaptation. A large amount of data had already been collected on this subject. However, Held and Hein had an unanswered question: does an individual need to physically move themselves around when they experience visual changes in order to develop their perceptual skills?

They decided to test whether being moved around (by someone or something else) while experiencing visual changes was enough to develop visually-guided behaviour, or whether the individual needed to move themselves around in order to learn such skills.

Richard Held

Sample The sample was made up of 20 kittens organised into ten pairs (meaning two kittens in one pair… please refrain from becoming too impressed with my mathematical skills). In each pair, one kitten was labelled ‘active’ (A) and the other ‘passive’ (P). Eight pairs were named Group X and were brought up in darkness until they were 8-12 weeks old, then they were exposed to the apparatus. Two pairs were named Group Y and were exposed to the patterned interior of a laboratory while restrained from 2-10 weeks of age for three hours a day.

Procedure A roundabout apparatus was used. It was built into a cylinder with black, white and “metal” coloured stripes on the inside walls. The centre of the roundabout was also striped.

Each pair of kittens was attached to a roundabout in two separate positions: Kitten A could move up, down, left, right, towards or away from the centre, rotate clockwise or anti-clockwise; Kitten P was carried in a basket so it could not move at all, only look and feel the floor with its paws. The centre of the roundabout prevented kittens from looking at each other and they could not look at their own limbs either. A mechanical system resulted in Kitten A’s movements being replicated onto Kitten P (if Kitten A moved left, Kitten P would also be moved to the left).

After that, the kittens’ perceptual skills were tested in a brightly lit laboratory. Three tests were administered:

• Visually-guided paw placement The kitten was held by an experimenter. Its head and forelegs hung freely. It was carried down to the edge of a table. A kitten with normal visual experience would extend its paws in anticipation of contact with the table.
• Avoidance of a visual cliff The kitten was placed onto a transparent cliff-shaped block. It was placed onto the centre. One side looked ‘shallow’ and the other looked ‘deep’. A kitten with normal visual experience completely avoids the deep-looking side because they have a perception of different depths.
• Blink to an approaching object The kitten was held in a standing position. The experimenter quickly brought his hand towards the kitten’s face, stopping just in front of it. A kitten with normal visual experience would blink in response.

Then, three further tests on visual receptors and responses were carried out.

• Visual pursuit of a moving object The experimenter moved their hand slowly in front of the kitten. The kitten’s eye movements were recorded. A kitten with normal visual experience would follow the hand with its eyes.
• Pupillary reflex to light A torch beam was moved across the kitten’s eyes. Any change in pupil size was noted. The pupil of a kitten with normal visual experience would shrink in response to the light.
• Tactual placing response The kitten was held with its forelegs hanging freely. Its front paws were placed against the vertical surface of a table. A kitten with normal visual experience would respond by moving its paws into a horizontal position on the table surface.

After these tests, the kittens were let loose in an illuminated room for 48 hours and retested.

Results All the ‘active’ kittens had developed the following:

• a normal visually-guided paw placement response
• a normal blinking response
• a normal level of depth perception
• a normal response to following movements

The ‘passive’ kittens had not developed any of these perceptual abilities; they would not blink in response to approaching objects, they had no sense of depth perception, they did not adjust their body positions. This supports the idea of self-generated movement alongside visual experience being necessary for the development of perception and visually-guided behaviour.

After the 48 hours spent in a bright room, the results of the tests showed that all kittens – whether active or passive – had developed a normal set of responses, reactions and perceptions.

Type of research method This was a laboratory experiment with scientific apparatus and controls.

Independent variable The IV was the condition of the kitten: active or passive.

Dependent variable The DV was the perceptual skills and development of the kittens.

• Very high level of control: The biggest strength is the amount of control exercised, for example: kittens were matched on the speed of travel, direction of travel, distance travelled, height from floor, contact with the floor, view of apparatus and surroundings, and the environment they were raised in.
• Replicable: Since the procedure was basically standardised, it is possible for us to replicate this experiment and check the reliability of the results.
• Not generalisable: As awesome as cats undoubtedly are, they do not represent all creatures on earth. They are mammals, like human beings, but it is questionable as to what the results would have shown if this experiment was conducted on humans or even on dogs.
• Questionable validity: It is possible that rather than proving that the perceptual abilities were learned, the experiment only distorted the inborn (natural) abilities of the kittens. In other words, maybe the kittens were born with these skills but the experiment affected them, instead of the kittens learning their skills from the experiment itself.
• Lack of ecological validity: Due to the artificial conditions of the experiment, the results aren’t likely to be as relevant for real life situations. Kittens aren’t normally bred in dark laboratories and most of them don’t spend their days on miniature roundabouts.

Ethical issues

• Informed consent: As far as I know, animals (even clever ones like kittens) cannot and do not give consent.
• Deception: Kittens were not and probably cannot be deceived.
• Confidentiality: Don’t worry, nobody knows the real identities of any of the kittens used in this study.
• Emotional or physical harm: Time to get serious, this ethical issue may actually be relevant. The experimenters basically kept these kittens in complete darkness from birth and put them on little carousels for three hours a day. This may have hindered or disturbed the kittens emotionally, as well as physically distorting their development. Although the kittens eventually showed “normal” responses and skills, they were not living like normal kittens.
• The right to withdraw: Animals cannot exercise this right.
• Debriefing: As fun as it is to sit and have long, deep, intellectual conversations with our feline friends, debriefing is not something that they usually get to take part in.

Reference: Held, R. and Hein, A. (1963). Movement-Produced Stimulation in the Development of Visually Guided Behavior. Journal of Comparative and Physiological Psychology. 56 (5): 872-876.

Share this:

21 thoughts on “ development of visually guided behaviour ”.

Hello! I would love to use the first image of the kittens in the carousel for a research paper I am writing for a class. Please let me know where you got this image from or if I may have your permission to use it! Thank you for your time!

Pingback: ShortCutstv | Psychological Studies Deconstructed

Does anyone know why the apparatus had stripes? -M xo

My saviour 🙂

Can I ask what was the purpose of group Y? I know they are a control group but how do I explain that further? and also why was the Passive Kitten in Group Y exposed to the carousel for another 126 hours before being tested? Thanks so much for the notes btw 🙂

Hey, can you help me with an alternative method of held and hein to investigate visually guided behaviour? For example by using a sample other than animals? I have a college exam in 4 days. Please reply as soon as possible. Thanks!

Thank you sooooooo muchh! Bless you!!

I just love your sense of humour! 😀 “Long, deep, intellectual conversations with our feline friends” 😂

Hey, I want to know an alternative method for Held and Hein! Please get back to me asap

Hi Mishaal,

Sorry but this question is SO vague! You need to be more specific. Alternative method for replication or alternative method to investigate visually-guided behaviour? Are you looking at a past paper question? There are actually so many possibilities.

Thanks, Maryam

Hi Maryam ,

If it’s a alternative study for replication how should I do it ? Thank you (:

Hey Maryam.. Why were kittens used in this experiment and not any other non-human animals?

As kittens are “higher mammals”, like humans, they were a good choice as undoubtedly Held and Hein wanted their sample to be as generalisable to humans as possible to be of any use.

I have a question… Why didn’t Held & Hein take sound into consideration? Like, when the kittens were in the apparatus, they could not see each other, so what if the kittens meowed? They would know that another kitten is present there.. Does sound have any effect on the spatial perception of kittens?

Ok nevermind I got my answer

Heyy, I just wanted to ask whether you could incorporate some statistics in the summaries of the core studies because my teacher says it gives an added advantage when it is being marked???

Do you mean general statistics? If I’m not wrong, statistics are only relevant in certain places, such as the results (e.g. 60% of participants showed…). In that case, I’ve mentioned it on most of the studies (usually in the Results section). I’ll take another look through all the studies and see if I’ve missed anything out or if I can add more! Feel free to expand on this if I am missing anything! Would appreciate any prods in the right direction, haha.

Thanks for your suggestion and comment!

Best wishes, Maryam XXX

why are the kittens being kept in dark cages ?

If I’m not wrong, this was a controlled variable. The kittens were all put into a ‘lit-up’ environment at the same time; before that, they were all in the same dark environment. If anyone thinks this needs correcting, please do so!

Hope that helps, Maryam XXX

they were reared in the dark until the age of 10-12 weeks to identify the answer to the nature nurture debate, that despite being reared in the darkness, and after exposure only the active kittens acquire the perception but the passive kittens did not indicates that the behavior is nurture and not nature. and then the sensory rearrangement technique answers the question that if nurture is required then simple visual stimulation is not sufficient but freedom of movement along with visual stimulation is a necessary requirement. hope this explains you the purpose of that control.

Hi I want to use the lower image of the kitten carousel in a book I’m writing – can you tell me, did you get permission and if so who from? Thanks. Guy Claxton – please reply to [email protected]

Leave your comments below! Cancel reply

• Copy shortlink
• Report this content
• Manage subscriptions

Reference Library

Collections

• See what's new
• All Resources
• Student Resources
• Assessment Resources
• Teaching Resources
• CPD Courses
• Livestreams

Study notes, videos, interactive activities and more!

Psychology news, insights and enrichment

Currated collections of free resources

Browse resources by topic

• All Psychology Resources

Resource Selections

Currated lists of resources

Study Notes

Blakemore and Cooper (1970)

Last updated 22 Mar 2021

• Share on Facebook
• Share on Twitter
• Share by Email

Impact of Early Visual Experience. Area – Biological Psychology. Key Theme: Brain Plasticity

Background and aim: The aim of this experiment was to investigate the physiological and behavioural effects of a limited visual experience and whether brain development/plasticity occurs due to experiences rather than nature.

Method: This study was a laboratory experiment , taking place in a controlled artificial environment. The participants in the study were new born kittens who were immediately placed into a dark room. At two weeks of age the kittens were then randomly placed into one of two conditions for five hours a day: this was either a horizontal or a vertical environment ( IV ).

The kittens had to stand on a clear glass platform which was inside a tall cylinder of which the inner surface was covered with either horizontal or vertical black-and-white stripes. Additionally, there were no corners or edges in their environment. The kittens’ visual field was restricted to 130 degrees, as they were required to wear a wide black collar. This prevented them from seeing their own body and ‘beyond their world of stripes’. Blakemore and Cooper defended the ethics of the study by stating that the kittens did not seem distressed, as they inspected the walls of the tube for long periods of time.

After five months, exposure to the experimental conditions ended and the kittens were then placed for several hours a week from their dark cage to a small, well-lit furnished room. The Dependent Variable ( DV ) was then measured: this was whether kittens raised in a horizontal environment could detect vertically aligned objects and vice-versa. After 7 and a half months, two of the kittens, one from each environment were anaesthetised and their neurophysiology was examined.

Results: All the kittens were extremely visually impaired; they demonstrated no visual placing when brought up to a table top and had no startle response when an object was thrust towards them. However, their papillary reflexes were normal and they guided themselves mainly by touch. All kittens showed behaviour blindness, meaning that they could not detect objects or contours that were aligned in the opposite way to their previous environment. They also demonstrated fear when they were standing on the edge of a surface.

There was recovery of some deficiencies from their early deprivation. After about 10 hours the kittens showed visual placing and some startled responses; they could also easily jump from a chair to the floor. But the kittens did suffer some permanent damage such as trying to touch things well beyond their reach and following objects with clumsy head movements. Moreover, the neurophysiological examination found no evidence of astigmatism (blurred vision), but there was evidence that horizontal plane recognition cells did not ‘fire-off’ in the kitten from the vertical environment and vice-versa – meaning that kittens were unable to perform orientation selectivity and therefore they suffered from ‘physical blindness’.

Research method: As this was a laboratory experiment, the kittens’ environments were highly controlled and therefore causal conclusions can be made. The study has levels of internal validity - we can infer that the IV (environment) caused visual impairment and neurophysiological damage (DV). The study could also be easily replicated in order to test the reliability of the findings (although this wasn’t done).

Ethical considerations: Exposing animals to a dark room for two weeks and then up until 5 months of age a visually depriving environment, could be considered to be psychologically harmful for the kittens. However, Blakemore and Cooper reported no distress from the animals. Furthermore, the study complied with the ethical guidelines for animal research. It could also be argued that any harm to the animals were outweighed by the usefulness of this research.

Sampling bias: Blakemore and Cooper would argue that due to some physiological similarities between cats and humans, we can generalise results to humans. However, critics would argue against this due to obvious differences between the species. Furthermore, as the sample was very small (only 2 cats’ neurophysiology was examined) we may not be even able to generalise to other cats!

For More Study Notes…

To keep up-to-date with the tutor2u Psychology team, follow us on Twitter @tutor2uPsych , Facebook AQA / OCR / Edexcel / Student or subscribe to the Psychology Daily Digest and get new content delivered to your inbox!

You might also like

Maguire et al. (2000), ib psychology (bloa): cognitions, emotions & behaviours are products of the anatomy and physiology of our nervous and endocrine system, plasticity & functional recovery: biopsychology webinar video.

Topic Videos

Biopsychology: Plasticity and Functional Recovery

15 must-see paper 2 biopsychology videos for a level psychology students.

9th June 2017

Example Answer for Question 6 Paper 2: A Level Psychology, June 2017 (AQA)

Exam Support

Biopsychology: MCQ Revision Test 1 for AQA A Level Psychology

Psychology in the News | Does Exercise Benefit the Brain?

29th January 2024

Our subjects

• › Criminology
• › Economics
• › Geography
• › Health & Social Care
• › Psychology
• › Sociology
• › Teaching & learning resources
• › Student revision workshops
• › Online student courses
• › CPD for teachers
• › Livestreams
• › Teaching jobs

Boston House, 214 High Street, Boston Spa, West Yorkshire, LS23 6AD Tel: 01937 848885

• › Contact us
• › Terms of use
• › Privacy & cookies

© 2002-2024 Tutor2u Limited. Company Reg no: 04489574. VAT reg no 816865400.

Attention and the depth perception of kittens

• Published: 22 November 2013
• Volume 26 , pages 248–251, ( 1988 )

Cite this article

• Richard D. Walk 1 ,
• Jane D. Shepherd 1 &
• David R. Miller 1  

705 Accesses

6 Citations

1 Altmetric

Explore all metrics

An experiment by Held and Hein (1963) in which active kittens discriminated depth and passive ones did not is often cited as showing the importance of self-produced locomotion for perceptual development. We hypothesized that their passive kittens may have paid less attention than active kittens to the environment. We repeated their experiment with kittens that either were given an interesting display to watch or were allowed to move through the environment by lifting their heads to close a microswitch that operated a go-cart. Kittens that paid greater attention to the environment discriminated depth on the visual cliff, whereas those reared with similar light exposure conditions but without the increased attention did not discriminate.

Article PDF

Download to read the full article text

Similar content being viewed by others

Binocular Vision and Depth Perception

What Is Attention?

Avoid common mistakes on your manuscript.

Berenthal, B. I, & Campos, J. J. (1987). New directions in the study of early experience. Child Development , 58 , 560–567.

Article   Google Scholar  

Berenthal, B. L, Campos, J. J. , & Barrett, K. C. (1984). Self-produced locomotion: An organizer of emotional, cognitive, and social development in infancy. In R. Emde & R. Harmon (Eds.), Continuities and discontinuities in development (pp. 175–210). New York: Plenum Press.

Chapter   Google Scholar  

Gibson, J. J. (1966). The senses considered as perceptual systems . Boston: Houghton Mifflin.

Google Scholar  

Gibson, J. J. (1979). The ecological approach to visual perception . Boston: Houghton Mifflin.

Hein, A. , & Held, R. (1967). Dissociation of the visual placing response into elicited and guided components. Science , 158 , 390–392.

Article   PubMed   Google Scholar  

Held, R. (1965). Plasticity in sensory-motor systems. Scientific American , 213 (5), 84–94.

Held, R. , & Hein, A. (1963). Movement produced stimulation in the development of visually guided behavior. Journal of Comparative & Physiological Psychology , 56 , 872–876.

Lackner, J. R. (1981). Some aspects of sensory-motor control and adaptation in man. In R. D. Walk & H. L. Pick, Jr. (Eds.), Intersensory perception and sensory integration (pp. 143–173). New York: Plenum Press.

Walk, R. D. (1966). The development of depth perception in animals and human infants. Monographs of the Society for Research in Child Development , 31 (107), 82–108.

Walk, R. D. (1969). Two types of depth discrimination by the human infant with five inches of visual depth. Psychonomic Science , 14 , 253–254.

Welch, R. B. (1978). Perceptual modification: Adapting to altered sensory environments . New York: Academic Press.

Download references

Author information

Authors and affiliations.

Department of Psychology, George Washington University, Washington, D.C., 20052

Richard D. Walk, Jane D. Shepherd & David R. Miller

You can also search for this author in PubMed   Google Scholar

Additional information

This experiment was supported in part by Grant MH-25,864 from the National Institute of Mental Health and by Biomedical Research Support Grant 2-S07-RR07019-14 National Institutes of Health to George Washington University.

Rights and permissions

Reprints and permissions

About this article

Walk, R.D., Shepherd, J.D. & Miller, D.R. Attention and the depth perception of kittens. Bull. Psychon. Soc. 26 , 248–251 (1988). https://doi.org/10.3758/BF03337301

Download citation

Received : 02 September 1987

Published : 22 November 2013

Issue Date : September 1988

DOI : https://doi.org/10.3758/BF03337301

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Human Infant
• Depth Perception
• Visual Depth
• Active Kitten
• Passive Group
• Find a journal
• Publish with us
• Track your research

In Cats David Hubel and Torsten Wiesel’s experiments showed that if a kitten is deprived of normal visual experience during a critical period at the start of its life, the circuitry of the neurons in its visual cortex is irreversibly altered.

In each of a number of newborn kittens, one eyelid was sutured shut. The kitten was allowed to grow up that way, and when it reached adulthood (around 6 months), its eyelid was opened again. Recordings were than made of the electrophysiological activity in each of the kitten’s eyes. These recordings showed an abnormally low number of neurons reacting in the eye that had been sutured shut, and an abnormally high number in the other eye. Macroscopic observation of the visual cortex showed that the ocular dominance columns for the eye that had been left open had grown larger, while those for the eye that had been closed had shrunk. Remarkably, Hubel and Wiesel also found that if the eye of an adult cat was sutured shut for a year, the responses of the cells in its visual cortex remain identical in all respects to those of a normal cat. Later experiments showed that suturing a cat’s eye shut had no effect on its visual cortex unless this visual deprivation took place during the first three months of the cat’s life. In Primates Other experiments have shown that this same phenomenon also occurs in primates, though the critical period is longer (up to age 6 months). Austin Riesen reared monkeys in darkness for the first 3 to 6 months of their lives. When these animals were then introduced into a normal environment, they had great difficulties in distinguishing even the simplest shapes. It took them weeks or even months to learn how to tell a circle from a square—a task that a normal monkey learns in a few days. Wiesel and Hubel also explored what happens in a monkey’s primary visual cortex when one of its eyelids is sutured shut for the first 6 months of its life. Normally, in monkeys, as in cats and humans, the two eyes work together to provide a single, three-dimensional image of the outside world, but this image is actually composed of two separate, slightly offset images on the two retinas. Wiesel and Hubel showed that it is not until the signals from the retinas reach the primary visual cortex that the brain begins to merge them into one, three-dimensional image. In monkeys who had one eyelid sutured shut right after birth, when the eyelid was opened again at 6 months of age, the animals had lost practically all useful vision in the eye that had been sensorily deprived. Yet recordings of electrophysiological activity in the ganglion cells of the retina of that eye, and the lateral geniculate nucleus cells for that eye, showed that these cells’ visual fields were normal and functional. It was only the primary visual cortex cells for that eye that showed practically no activity. Other experiments in which both eyelids were temporarily sutured shut showed that normal development of connectivity in the visual cortex does not depend on the absolute activity of the neural pathways from the two eyes, but rather on competition between the relative activities of these two pathways. As in other development processes for which there is a critical period, sensory deprivation does not have the same effect on adult animals—i.e., suturing shut one eyelid of an adult animal has no effect on the response of the visual cortex cells for that eye or for the other eye. In contrast, during the most sensitive part of the critical period, visual deprivation for as little as one week can have catastrophic effects on the animal’s vision for the rest of its lifetime.

In Humans In humans, certain diseases can cause a cataract (a total or partial opacity of the lens) in one or both eyes. Cataracts can occur not only in adults but also in very young children. Cataracts can now usually be removed surgically. Studies of individuals who had such surgery at various times in their lives showed that in humans as in other animals, there is a critical period for the development of the sense of sight. These studies demonstrated for the first time that early environmental influences, and hence particular neural activity patterns during a critical period, can permanently alter the neural connections in certain areas of the human brain.

• Neuroscience

The Cat Nobel Prize Part I

The surprising role that cats have played in the history of neuroscience..

Posted April 14, 2014

( Note: This is part of a series of posts I’m writing in conjunction with my new book, The Tale of the Dueling Neurosurgeons .)

Guinea pigs are practically synonymous with experiments. Lab rats have become the workhorses of modern medicine. Genetics owes a huge debt to the humble fruit fly. There’s almost no branch of the life sciences, in fact, that hasn’t leaned heavily on one animal or another. Even psychologists—who study the human mind in all its uniqueness—have employed chimpanzees in studies, teaching them sign language or raising them as “children” to see how they handle suburbia. (Not well...)

The cat has received far less fanfare as a model organism, possibly because we don’t like to think about using house pets in experiments. And truth be told, reading about some of the experiments that cats have endured can stir up real discomfort. But it’s at least some solace that Felis domesticus has played a crucial role in several triumphs of twentieth-century neuroscience , and I’ve decided to honor their sacrifice with a trio of posts. Here goes.

The first breakthrough, by David Hubel and Torsten Wiesel in the 1950s, involved sight. Hubel and Wiesel were studying a part of the brain known as the primary visual cortex (in the very back of the brain, near the bump on your noggin). In particular they wanted to know how neurons there responded to different shapes, and for whatever reason—possibly because cats have large, well-developed vision centers—they chose to work on cats. Frankly, the experiments conjure up thoughts of A Clockwork Orange : they strapped each cat down in a chair with its eyelids propped open and forced it to stare at a projection screen. (Thankfully, the cats were anesthetized and felt no pain.) Wires that measured neuronal activity ran into the cat’s brain, and Hubel and Wiesel projected images onto the screen to see what got the neurons excited.

At first, nothing seemed to excite the neurons: they simply wouldn’t fire, and the duo wasted months on fruitless experiments. Finally, though, after a slide got jammed in the projector, one neuron went off like a machine gun: rat-a-tat-tat-tat. (This was a total mistake—and one of the best examples of serendipity in science history. You can read the full, hilarious story here.) After some desperate fiddling, Hubel and Wiesel realized that the neuron was responding to the sharp shadow that the edge of the slide was making on the screen. This neuron, in other words, dug lines.

Follow-up work revealed that this was in fact a general principle of vision neurons: no matter what shape our eyes are looking at—a spiral nautilus, the curve of a hip—when the input reaches our primary visual cortex, the neurons there determinedly break that image down into line segments. It’s as if everything we see around us is stick figures. Before these cats, no one had even suspected this possibility.

These clockwork kitties revealed something else critical as well: that we see moving images much more easily than still images. Why? Because, evolutionarily speaking, it’s probably more critical for animals like cats to spot moving things (dogs, mice, falling trees) than static things, which can wait. In fact, animal vision—including human vision—is so biased toward movement that we don’t technically see stationary objects at all. To see something stationary, our brains have to scribble our eyes very subtly over its surface so that it appears to move, at least a little. (Experiments by later scientists, who built on Hubel and Wiesel’s work, have proved that if you artificially stabilize an image on the retina with a combination of special contact lenses and microelectronics, the image vanishes completely.)

Overall, some observers claimed that neuroscientists learned more about vision during a few years of work by Hubel and Wiesel than in the previous two centuries, and they won a richly deserved Nobel Prize in 1981. The duo wasn’t done, either. The next post in this series explores how other, even more put-upon cats helped reveal how our brains learn to see in the first place during infancy.

(Coming up in part II: Kittens, stripes, and “critical windows” in brain development.)

Sam Kean is a science writer and the author of The Violinist's Thumb and The Disappearing Spoon.

• Find a Therapist
• Find a Treatment Center
• Find a Psychiatrist
• Find a Support Group
• Find Online Therapy
• United States
• Brooklyn, NY
• Chicago, IL
• Houston, TX
• Los Angeles, CA
• New York, NY
• Portland, OR
• San Diego, CA
• San Francisco, CA
• Seattle, WA
• Washington, DC
• Asperger's
• Bipolar Disorder
• Chronic Pain
• Eating Disorders
• Passive Aggression
• Personality
• Goal Setting
• Positive Psychology
• Stopping Smoking
• Low Sexual Desire
• Relationships
• Child Development
• Self Tests NEW
• Therapy Center
• Diagnosis Dictionary
• Types of Therapy

It’s increasingly common for someone to be diagnosed with a condition such as ADHD or autism as an adult. A diagnosis often brings relief, but it can also come with as many questions as answers.

• Emotional Intelligence
• Gaslighting
• Affective Forecasting

Roger Sperry’s Split Brain Experiments (1959–1968)

Editor's note: Sarah Walls created the above image for this article. You can find the full image and all relevant information here .

In the 1950s and 1960s, Roger Sperry performed experiments on cats, monkeys, and humans to study functional differences between the two hemispheres of the brain in the United States. To do so he studied the corpus callosum, which is a large bundle of neurons that connects the two hemispheres of the brain. Sperry severed the corpus callosum in cats and monkeys to study the function of each side of the brain. He found that if hemispheres were not connected, they functioned independently of one another, which he called a split-brain. The split-brain enabled animals to memorize double the information. Later, Sperry tested the same idea in humans with their corpus callosum severed as treatment for epilepsy, a seizure disorder. He found that the hemispheres in human brains had different functions. The left hemisphere interpreted language but not the right. Sperry shared the Nobel Prize in Physiology or Medicine in 1981for his split-brain research.

Sperry also studied other aspects of brain function and connections in mammals and humans, beyond split-brains, in 1940s and 1950s. In 1963, he developed the chemoaffinity hypothesis, which held that the axons, the long fiber-like process of brain cells, connected to their target organs with special chemical markers. This explained how complex nervous systems could develop from a set of individual nerves. Sperry then also studied brain patterns in frogs, cats, monkeys, and human volunteers. Sperry performed much of his research on the split-brain at California Institute of Technology, or Caltech, in Pasadena, California, where he moved in 1954.

Sperry began his research on split-brain in late 1950s to determine the function of the corpus callosum. He noted that humans with a severed corpus callosum did not show any significant difference in function from humans with intact corpus callosum, even though their hemispheres could not communicate due to the severing of the corpus callosum. Sperry postulated that there should be major consequences from cutting the brain structure, as the corpus callosum connected the two hemispheres of the brain, was large, and must have an important function. Sperry began designing experiments to document the effects of a severed corpus callosum. At the time, he knew that each hemisphere of the brain is responsible for movement and vision on the opposite side of the body, so the right hemisphere was responsible for the left eye and vice versa. Therefore, Sperry designed experiments in which he could carefully monitor what each eye saw and therefore what information is was going to each hemisphere.

Sperry experimented with cats, monkeys, and humans. His experiments started with split-brain cats. He closed one of their eyes and presented them with two different blocks, one of which had food under it. After that, he switched the eye patch to the other eye of the cat and put the food under the other block. The cat memorized those events separately and could not distinguish between the blocks with both eyes open. Next, Sperry performed a similar experiment in monkeys, but made them use both eyes at the same time, which was possible due to special projectors and light filters. The split-brain monkeys memorized two mutually exclusive scenarios in the same time a normal monkey memorized one. Sperry concluded that with a severed corpus callosum, the hemispheres cannot communicate and each one acts as the only brain.

Sperry moved on to human volunteers who had a severed corpus callosum. He showed a word to one of the eyes and found that split-brain people could only remember the word they saw with their right eye. Next, Sperry showed the participants two different objects, one to their left eye only and one to their right eye only and then asked them to draw what they saw. All participants drew what they saw with their left eye and described what they saw with their right eye. Sperry concluded that the left hemisphere of the brain could recognize and analyze speech, while the right hemisphere could not.

In the 1960s when Sperry conducted his split-brain research on humans, multiple scientists were studying brain lateralization, the idea that one hemisphere of the brain is better at performing some functions than the other hemisphere. However, researchers did not know which tasks each side of the brain was responsible for, or if each hemisphere acted independently from the other.

Sperry describes his research in cats in the article "Cerebral Organization and Behavior" published in 1961. To test how the cutting of the corpus callosum affected mammals, Sperry cut the corpus callosum of multiple cats and had them perform some tasks that involved their vision and response to a visual stimulus. After severing each cat´s corpus callosum, he covered one of the cat´s eyes to monitor with which eye the cat could see. Sperry could switch the eye patch from one eye to the other, depending on which visual field he wanted the cat to use. Next, Sperry showed the cats two wooden blocks with different designs, a cross and a circle. Sperry put food for the cat under one of the blocks. He taught the cats that when they saw the blocks with one eye, for instance, the right eye, the food was under the circle block, but when they saw it with the left eye, the food was under the block with a cross. Sperry taught the cats to differentiate between those two objects with their paws, pushing the correct wooden block away to get the food.

When Sperry removed the eye patch and the cats could see with both eyes, he performed the same experiment. When the cats could use both eyes, they hesitated and then chose both blocks almost equally. The right eye connects to the left hemisphere and the left eye connects to the right hemispheres. Sperry suspected that since he cut the corpus callosum in those cats, the hemispheres could not communicate. If the hemispheres could not communicate and the information from one eye only went to one hemisphere, then only that hemisphere would remember which block usually had food under it. From that, Sperry concluded that the cats remembered two different scenarios with two different hemispheres. He suspected that the cats technically had two different brains, as their hemispheres could not interact and acted as if the other one did not exist.

Sperry performed a similar experiment with monkeys, in which he also cut their corpus callosum. He wanted to test if both hemispheres could operate at the same time, even though they were not connected. That required separation of visual fields, or making sure that the right eye saw a circle, while the left eye saw a cross, like in the cat experiment, but without an eye patch and both eyes would see something at the same time instead of interchanging between the open eyes. Sperry solved that by using two projectors that were positioned side-by-side at an angle and showed mutually exclusive images. For example, the projector on the right showed a circle on the left and a cross on the right, while the projector on the left showed a cross on the left and a circle on the right. Sperry placed special light filters in front of each of the monkey´s eyes. The light filters made it so that each eye saw the images from only one of the projectors. That meant one of the eyes saw the circle on the right and the cross on the left, while the other eye saw the cross on the right and the circle on the left. From his experiments with cats, Sperry knew that there was no sharing of information from right and the left hemispheres, so he made the monkeys memorize two different scenarios at the same time.

The left eye saw a scenario where food would be dispersed when the monkey pressed the button corresponding to a cross, while the right eye saw a scenario where food would be dispersed when the monkey pressed a button corresponding to a circle. Ultimately, it was the same button, but the eyes saw it differently because of two projectors and special light filters. Sperry concluded that both hemispheres of the brain were learning two different, reversed, problems at the same time. He noted that the split-brain monkeys learned two problems in the time that it would take a normal monkey to learn one, which supported the assumption that the hemispheres were not communicating and each one was acting as the only brain. That seemed as a benefit of cutting corpus callosum, and Sperry questioned whether there were drawbacks to the procedure.

Sperry performed the next set of experiments on human volunteers, who had their corpus callosum severed previously due to outside factors, such as epilepsy. Sperry asked volunteers to perform multiple tests. From his previous experiments with cats and monkeys, Sperry knew that one, the opposite, hemisphere of the brain would only analyze information from one eye and the hemispheres would not be able to communicate to each other what they saw. He asked the participants to look at a white screen with a black dot in the middle. The black dot was the dividing point for the fields of view for a person, so the right hemisphere of the brain analyzed everything to the left of the dot and the left hemisphere of the brain analyzed everything that appeared to the right of the dot. Next, Sperry showed the participants a word on one side of the black dot for less than a second and asked them to tell him what they saw. When the participants saw the word with their right eye, the left hemisphere of the brain analyzed it and they were able to say what they saw. However, if the participants saw the word with their left eye, processed by right hemisphere, they could not remember what the word was. Sperry concluded that the left hemisphere could recognize and articulate language, while the right one could not.

Sperry then tested the function of the right hemisphere. He asked the participants of the same experiment that could not remember the word because it was in the left visual field to close their eyes and draw the object with their left hand, operated by the right hemisphere, to which he presented the word. Most people could draw the picture of the word they saw and recognize it. Sperry also noted that if he showed the word to the same visual field twice, then the person would recognize it as a word they saw, but if he showed it to the different visual fields, then the participants would not know that they saw the word before. Sperry concluded that the left hemisphere was responsible not only for articulating language, but also for understanding and remembering it, while the right hemisphere could only recognize words, but was not able to articulate them. That supported the previously known idea that the language center was in the left hemisphere.

Sperry performed another similar experiment in humans to further study the ability of the right hemisphere to recognize words. During that experiment, Sperry asked volunteers to place their left hand into a box with different tools that they could not see. After that, the participants saw a word that described one of the objects in the box in their left field of view only. Sperry noted that most participants then picked up the needed object from the box without seeing it, but if Sperry asked them for the name of the object, they could not say it and they did not know why they were holding that object. That led Sperry to conclude that the right hemisphere had some language recognition ability, but no speech articulation, which meant that the right hemisphere could recognize or read a word, but it could not pronounce that word, so the person would not be able to say it or know what it was.

In his last series of experiments in humans, Sperry showed one object to the right eye of the participants and another object to their left eye. Sperry asked the volunteers to draw what they saw with their left hand only, with closed eyes. All the participants drew the object that they saw with their left eye, controlled by the right hemisphere, and described the object that they saw with their right eye, controlled by the left hemisphere. That supported Sperry´s hypothesis that the hemispheres of brain functioned separately as two different brains and did not acknowledge the existence of the other hemisphere, as the description of the object did not match the drawing. Sperry concluded that even though there were no apparent signs of disability in people with a severed corpus callosum, the hemispheres did not communicate, so it compromised the full function of the brain.

Sperry received the 1981 Nobel Prize in Physiology or Medicine for his split-brain research. Sperry discovered that the left hemisphere of the brain was responsible for language understanding and articulation, while the right hemisphere could recognize a word, but could not articulate it. Many researchers repeated Sperry´sf experiments to study the split-brain patterns and lateralization of function.

• Sperry, Roger W. "Cerebral Organization and Behavior." Science 133 (1961): 1749–57. http://people.uncw.edu/puente/sperry/sperrypapers/60s/85-1961.pdf (Access December 8, 2017).
• Sperry, Roger W. "Hemisphere Deconnection and Unity in Conscious Awareness." American Psychologist 28 (1968): 723–33. http://people.uncw.edu/Puente/sperry/sperrypapers/60s/135-1968.pdf (Access December 8, 2017).
• Sperry, Roger W. "Split-brain Approach to Learning Problems." In The Neurosciences: A Study Program , eds. Gardner C. Quarton, Theodore Melnechuk, and Francis O. Schmitt, 714–22. New York: Rockefeller University Press, 1967. ttp://people.uncw.edu/puente/sperry/sperrypapers/60s/130-1967.pdf (Accessed November15, 2017).
• "The Split Brain Experiments." Nobelprize.org . https://www.nobelprize.org/educational/medicine/split-brain/background.html (Accessed May 3, 2017).

How to cite

Articles rights and graphics.

Copyright Arizona Board of Regents Licensed as Creative Commons Attribution-NonCommercial-Share Alike 3.0 Unported (CC BY-NC-SA 3.0)  

Last modified

Share this page.

Zing-Yang Kuo (Part 1)

At this early stage of the 21st century, we find our discipline at a watershed. Developments in molecular genetics, developmental and comparative embryology, and neuroscience over the last few decades have greatly enhanced our understanding of how biology is related to behavior as a set of participating and not causative factors. Advances in conceptual and theoretical frameworks have allowed for the beginning of a synthesis of biological-level factors with interpersonal and social-ecological factors. Coupling these theoretical advances with increased computing power and powerful analytic methodologies ranging from structural equation approaches to nonlinear dynamics and complex systems theory has resulted in a psychology at the precipice of a major paradigmatic shift.

Reading the works of Zing-Yang Kuo in the context of these times for psychology, we refer to the words of a learned professor to his young graduate student: “If you want new ideas, read old books,” for Kuo’s work in many instances presages the cutting-edge advances in developmental and comparative psychology. Indeed, much of his writing is so contemporary and salient one may forget that much of it was written a century ago. Kuo began his career at a time when biology and psychology were in their infancy and flush with new and competing philosophical and theoretical ideas. In biology, the full force of Darwinian evolution by descent was taking shape and the rediscovery of Mendel’s work was leading towards the neo-Darwinian modern synthesis. Prominent scientists such as Woodger (1929) and Needham (1929) were developing a uniquely biological philosophy. On the horizon were similar efforts by psychologists.

Kuo’s writing falls into two distinct periods. The first ranges from the early 1920's to the late 1930's. During this time, most of Kuo’s writings focused on general philosophical, theoretical, and methodological statements concerning the study of behavior. He also conducted his important studies of the prenatal development of chickens. The second period (appearing as “Part 2” in the next issue of this newsletter) falls between 1960 and 1967. During this latter period, Kuo developed three general theoretical principles fully discussed in his book, “Dynamics of Behavioral Development.” The three theoretical principles outlined in this latter book are the theory of behavioral gradients, the theory of behavioral potentials, and his organizing framework of behavioral epigenesist, which is today among the important topics not only in biology but in psychology and developmental science as well.

Reading Kuo, it is easy to get the impression that he was a behaviorist. Indeed, many of his criticisms of the psychology of his day were framed in a behaviorist framework. Thus, he argued forcefully for an objective science of behavior; he took strong positions against the introspection approach of structuralists; he went so far as to point out that while he was a student of Tolman’s he in no way adopted his cognitive way of thinking. Yet, even early in his career, Kuo became critical of behaviorism for not ridding itself of traditional psychological concepts and adhering to its objective ideals.

Kuo’s writing spans some five decades (1921-1970), beginning with the very important stage-setting paper of 1921, “Giving up instincts in psychology,” and concluding with his equally important book of 1967, “Dynamics of Behavior Development.” All the remaining of his 33 papers can be seen as steps leading directly from proposals and ideas first presented in 1921 and elaborated into one of the few systematic theories of general psychology of the 20th century. The last paper he wrote in 1970 was fittingly included in a paean to T. C. Schneirla, the only other comparative psychologist of the 20th century who was devoted to developing a unified theoretical position for comparative psychology. Kuo, and the psychologists he influenced, worked from an anti-hereditarian, developmental-contextual perspective that is only now being widely appreciated in the discipline.

Kuo’s anti-instinct position

Kuo began publishing in 1921 when psychology was a far different enterprise from what it is today. While scientific psychology was in its infancy in Kuo’s day, the exact nature of the scientific practices of the discipline had not been worked out in any consensus form by 1920. Indeed, it is possible to argue that it is only at the end of our first 100 years, at the beginning of this new century, that this still-young science has at last worked out the details of conducting psychology as a natural science. In 1920, psychology was very much influenced by the idea of instinct; that much of animal behavior was inherited, as was much of human behavior as well. Kuo’s first paper (1921) argued forcefully against this idea. Though he was certainly not the only anti-instinct psychologist writing at the time, he was arguably the most strident. There are several reviews of the gestation of the instinct concept, among them by Beach (1955) and Diamond (1973).

It comes as no surprise that Kuo could not find agreement on the definition of the instinct concept in 1920. Psychology has been plagued with this problem of defining its terms and concepts for all of its 120 year existence – is there yet agreement about the meanings of intelligence, mental illness, aggression? However, for argument’s sake, Kuo adopted a definition that seems to have persisted until the idea was formally described and defined by the Nobel laureates in ethology, Nikolas Tinbergen and Konrad Lorenz.

Citing Parmelee (no date), Kuo writes, “An instinct is an inherited combination of reflexes which have been integrated by the central nervous system so as to cause an external activity of the organism which usually characterizes a whole species and is usually adaptive (1921, p. 646; cited from Parmelee, “The Science of Human Behavior,” no date). That is, instincts are species specific, biologically adaptive, fixed and stereotyped, and triggered by specific stimuli only. Compare this to the definition of classical ethology of the mid-20th century, that instincts, as inherited behavior patterns, are: characteristic of the species they appear in; stereotyped and constant in form; behaviors which appear in animals isolated from conspecifics; and which appear without practice, fully formed when first triggered by appropriate sign stimuli (Hess, 1962).

Instinct theory, that behavior was inherited, was part of scientific psychology from its inception. Kuo tells us that though long relegated only to animals, it was William James who reintroduced the inheritance of behavior to human psychology. The impact of James’ chapter on Instinct” in the “Principles of Psychology” was enormous and many of his contemporaries expanded James’ ideas “with abandon,” according to Harry Harlow (1969). References at the time to instincts were not to reflexes, but rather to complex coordinated behaviors: climbing by children, fear of strange men, pugnacity and anger (James; Harlow, 1969); human character and the human mind (Thorndike; Kuo, 1921); business acumen, religion, and “every other affair of life” (McDougal; Kuo, 1921, p. 646). As Kuo points out in a remarkably contemporary sounding paragraph, human behavior, our social institutions, religious motivations, social unrest, and the labor movement “are to be explained in terms of instinct” (1921, p. 645). Some currently ascribe to the genes such complex human behavior such as television watching, divorce, what we like in a mate, murder, religion (Horgan, 1993, 1995).

Given Kuo’s conceptualization of psychology, it is not surprising that his very first publication was an argument against instinct theory. It is noteworthy that he wrote his first paper as a senior undergraduate at Berkeley, as astonishing an accomplishment for an undergraduate in 1921 as it is today (Gottlieb, 1976). His first argument against instinct would be elaborated in three additional papers (1922a, 1924, 1929b). Of course, these criticisms of instinct theory have meaning only in the context of Kuo’s definition of psychology as “the science which deals with the physiology of bodily mechanisms involved in the organismic adjustments to environment with special emphasis on the functional aspect of the adjustment” (1924, p. 427, italics in the original). The subject matter of psychology, behavior, what an animal does in response to a stimulus, “is solely physical and mechanical events.” (1924, p. 427)

Kuo posited several criteria for rejecting the hereditarian view in psychology:

• As we have already stated, there was no agreement in 1920 as to the precise definition of instinct
• The acquisition of so-called instincts. Kuo believed that a newborn infant acquires its diverse behavioral repertoire gradually through its life. During development, if the psychologist looks carefully, he will discover that factors other than instinct are responsible for the acquisition of all behavior
• Instincts imply purpose and teleology. As discussed below, Kuo rejected these concepts
• Methodological limitations of the genetic method (not genetics as we understand the term today, but rather an observational approach to development) lead to labeling a behavior an instinct merely because we observe it to occur within a species: “But, a careful analysis will show that the members of a species have similar reactions, not because they have inherited the same instincts, but rather, because they have inherited the same action system and live in a similar environment” (Kuo, 1921, p. 652)
• Instincts are biologically adaptive.

This is still a fundamental principle of hereditarian psychology. Evolutionary psychologists today propose that all behavior is adaptive, it being the result of Darwinian evolution. Just as cogent arguments are made today against this position (e.g., Gould, 1997a,b), so did Kuo: “It will be very ridiculous to say that the young infant attempts to grasp the fire or a harmful snake, when presented to him, because such a reaction is useful to the organism...To say that the so-called innate responses of the young human organism have biological value is to overlook the fact that from the moment that the child is born it is taken care of by society,” that is, provided a stimulus world within which to develop (1921, p. 654).

What, then, was Kuo’s alternative to the hereditarian approach? Crediting Watson for the idea, Kuo noted (1924) that the newborn comes into the world with a huge repertoire of individual motor reactions, the result of spontaneous neural and muscular activity and of random acts. These “units of reaction,” are muscular movements of parts of the body which occur prenatally. Every movement prior to birth occurs in an environment rich with stimuli and these stimuli are responsible for new motor acts, each of which then stimulate subsequent acts, etc. The newborn is extremely active and easily aroused, conditions which favor exploration and new stimulation, in turn leading to new behaviors. Except for vegetative acts (metabolic activities such as heart beat, gastric motility, breathing) no inborn behavior is adaptive or purposive. After birth, these motor acts are constantly organized, reorganized and integrated during the life history of the organism. Walking, for example, involves no new motor acts that the infant has not already engaged in. Rather, walking requires the coordination and integration of the legs, feet, head, trunk, eyes and more, as well as the maturation of the muscular and skeletal systems. All complex behavior can be understood this way as consisting of new organizations and integrations of old behavior units – organization upon organization upon organization. The new in a new habit is the new combination of old reaction units: “The development of human behavior is essentially the increase of complexity in the organization of reaction systems” (Kuo, 1921, p. 663). There is much that contemporary critics of hereditarian psychology will recognize in the writing of Kuo more than half a century ago. To be sure, we have learned a great deal since then, but his prescriptions remain the same. Little of what he had to say about the origins of behavior can be rejected today.

The genesis of the cat’s response to the rat

Not content to merely make anti-hereditarian propositions about behavioral origins, Kuo went on to provide empirical verification of his views. The first series of experiments he reported (1930, 1938a) addressed the question, “Is a cat a rat killer or rat lover?” Rat killing by cats meets all the criteria of an instinct. It is characteristic of all cats, is adaptive, stereotyped and specific to a particular stimulus – rats. It was not even denied by Kuo that some cats can do this without specific training. Thus, the purpose of his investigations was not to determine if training was necessary for rat killing, but, rather, to “manipulate the conditions in which the kitten is made to live so as to see what variations in its behavior toward the rat might be brought forth” (1930, p. 2).

Kittens were raised in nine conditions: 

• In isolation, kept from all animals after weaning
• In a rat-killing environment, with their rat-killing mothers
• In the same cage with different kinds of rats until 4 months of age
• As vegetarians and non-vegetarians; 5) Satiated or 12 hours food deprived at the rat-killing test
• Given specific rat-killing training
• Participating in rat-killing with other cats
• Tested with three species of rat to determine species preferences in killing It is curious that Kuo identified the three “rat” species as an albino rat, a wild gray rat, and a dancing mouse. This in no way impacts on the results of this research
• Training kittens to fear rats.

It is of interest to note that in reporting the results of all his experiments, Kuo used the simplest of statistical analyses, percentages and individual data; no inferential statistics, no null hypothesis testing. Kuo’s conceptions of behavior and resulting hypotheses are as complex and sophisticated as any in contemporary studies of behavior and seemingly incapable of being tested without equally sophisticated analytic techniques. However, in science, method is more important than statistical technique. Indeed, much of the sophistication of current statistical analysis is designed to address deficiencies in method. Kuo’s analysis of his results emphasize the fact that a clever experiment is often more valuable for answering a well thought out question, than it is for its complex statistics.

The main features of his results were: 

• In the first, isolation, experiment, 11 of 20 kittens did not kill rats when tested later. Kuo concluded, “Many psychologists believe that an instinct is universal in a species. But...11 kittens out of 20...did not seem to possess the rat-killing instinct” (p. 8)
• Kittens which watched their mothers kill rats, did so themselves and always killed the species they observed their mothers kill. Three of 18 kittens did not kill rats in this condition
• Of those kittens raised with rats, only three of 18 killed rats at the test, and they did not kill the species or individual they were raised with
• Kittens reared in isolation can be later trained to kill rats - nine of 11 - while those reared with rats apparently cannot be so trained - one of 15. Some kittens in this experiment ate rats even having never seen that occur, suggesting that rat eating can develop without any obvious reinforcement. The key term here is “obvious,” since Kuo always believed that some stimulus or experiential condition was at the foundation of every behavior
• Being raised on a vegetarian diet reduced rat eating, but not rat killing
• The size of the rat affects the age at which rat killing appears. Young kittens kill smaller rats earlier than they do larger ones
• Kittens reared with albino rats killed the other two species but not albinos
• All species of rats were killed, showing no innate tendencies to kill a particular species of rat
• Kittens kill and eat what they see their mothers kill and eat
• Kittens do not kill the species of rat they are raised with
• Larger rats evoke different responses from the kittens than do smaller ones. More hostile and negative behaviors were displayed to the larger than to the smaller rats, whereas smaller rats evoked playing behavior from the kittens
• 66 percent of all rat killing was done by kittens raised in the rat-killing environment; only 5 percent by kittens raised with rats
• Three kittens protected the rats they were raised with; some kittens apparently learned to “love” their rats (Kuo’s term)
• Only three of 16 kittens could be conditioned to fear rats; five kittens ran from the box they were shocked in during fear conditioning, but not from the rats
• Kittens raised not only with a rat, but in social grouping with two or three other kittens, were tolerant of rats, though not as attracted as they had been in the earlier one kitten-one rat rearing situation. On the other hand, all of the kittens seemed to form attachments to the other kittens they were raised with
• Kittens killed and ate hairless and shaved rat pups, but not unshaven pups
• Kittens raised for six months with sparrows ignored them for the first two months. They subsequently followed, caught, played with, and killed flying sparrows
• These same kittens reacted with indifference to sparrows in a garden setting.

This is an impressive set of conclusions, even by today’s standards. In one series of experiments Kuo was able to demonstrate the extreme malleability and plasticity of behavior that is characteristic of a species. The implication of this work is enormous. It shows quite clearly that an animal comes into the world with the potential to behave, if a cat, in typical cat fashion, or, depending on its experiential and stimulus history, in extremely atypical fashion. Can there be a stronger demonstration against the inheritance of behavior? In summing up this work, Kuo says, “To me, the organismic pattern...or bodily makeup and the size should be sufficient to tell why the cat behaves like cat, the tiger like tiger or the monkey like monkey. The cat has a cat body and hence the rat-killing behavior; the tiger has a tiger body, and hence man-killing behavior. The chimpanzee has a chimpanzee body, and so uses sticks and does things almost human....But the cat is a living machine; it grows and changes; it has a life history. Its behavior is being modified from the moment of fertilization to the point of death, and is modified according to the resultant forces of environmental stimulation, intra-organic as well as extra-organic. In other words, the kinds and range of potential responses of an organism are determined by its bodily size, and especially its bodily make-up or organismic pattern, while its actual responses are determined by its life history” (1930, p. 33). This is a crucial anticipatory statement of what Kuo will later refer to as the “principle of behavioral potentials” discussed in a later section of this paper.

Prenatal behavioral development

Kuo was a pioneer in the study of prenatal behavioral development, which even 30 years after his work still was little studied (Gottlieb, 1970). His efforts paved the way for later researchers in this area such as Gilbert Gottlieb (1971, 1973, 1976) and William Smotherman (Smotherman & Robinson, 1988, 1998). His findings on chicken embryo development (1938b; 1939, b, c) stand as an outstanding contribution to our understanding of the significance of prenatal events to later behavioral development. As we have pointed out in discussing Kuo’s criticisms of instinct, he believed an organism was born with a repertoire of reaction units which would become components of complex behaviors and which could be reorganized and integrated repeatedly, to form new complex behaviors. The only thing new about behaviors developed after birth, then, was in the combination and the sequence of these reaction units acquired during embryonic development. It is in this series of experiments that Kuo attempted to provide empirical verification of this idea, of the prenatal origins of behavioral units which would be subsequently available to be utilized functionally at birth.

Kuo said several times that the work of the psychologist begins where that of the instinct theorist ends. The hereditarian psychologist begins to examine behavior after the animal is born. But Kuo understood the great significance of prenatal events in influencing postnatal behavior. If every behavior had an experiential history, then the reaction units present at birth must also. How, though, to study prenatal events? Kuo believed that since the procedures used by the embryologists of his day presented an extremely unnatural environment, i.e., an extracted and incompletely formed embryo in a dissecting pan, the behaviors studied must also be unnatural (1932c). Thus, Kuo commented that Coghill’s studies of ambystoma suffered from making no reference whatever to the organism’s environment. And, of those studying mammals he said, “Those who have worked on the behavior of the mammalian fetus...removed the fetus from the uterus and gathered fragmentary information concerning its bodily movements...without noting the effect on behavior of the removal of the fetus from its normal fetal environment after the experimental delivery” (1932c, p. 245).

What was needed was a procedure which allowed observation and manipulation of the embryo without disturbing its normal environment. After experimenting with several techniques Kuo settled on a procedure which not only did not kill the embryos but also provided the least interference with normal functioning. A window was carefully cut into the shell of chicken eggs which was carefully peeled away so as not to disturb the underlying membrane. The membrane was painted with a thin layer of melted petroleum jelly which made the ordinarily opaque membrane rather clear, much as the glass in a shower door becomes clear with moisture on it. This allowed Kuo to observe some 3000 developing chicken embryos from fertilization to hatching. While he experimented on hens, pigeons, and ducks, he presents only data from chickens in these papers. He reported findings on ducks in a later collaboration with Gilbert Gottlieb (Gottlieb & Kuo, 1965). Except for obvious species differences (e.g., size, developmental timing) results with ducks mirrored those reported for chickens. Embryonic observation windows are shown in Gottlieb’s article elsewhere in this issue.

In selecting birds as a model for this series of experiments, Kuo noted that the environmental stimuli influencing the embryo’s activities was complex and included extra embryonic membranes, shell membranes, the egg shell, yolk and albumin, fluids in the extra-embryonic cavity. To study the development of the embryo outside the egg was to study it in an artificial environment completely devoid of these normally occurring environmental factors. The studies he conducted were laborious and the results extensive. He not only observed, he stimulated and measured reactions, some gross, some local and small, to compare his findings with those of others working with different species.

Prior to Kuo’s work, one could safely say that pecking for food by newly hatched chicks was an instinct. It is characteristic of the species, since all chickens do it; it is constant and stereotyped in form because all chickens do it the same way – they raise and lower their heads, open and close their beaks, and swallow, in a coordinated fashion; it appears without practice – the first time the chicken sees the grains it pecks at them, the grains acting as a specific or sign stimulus for the pecking behavior; and it appears in animals isolated from conspecifics (in their shells) and thus deprived of any opportunity to imitate this behavior. But, as Kuo undoubtedly believed, things are not always as they seem and patient, careful observation will always reveal some other factors at work. In the case of observations of a developing chicken embryo, what is to be seen is non-obvious, and so new techniques are required.

In looking at the developmental sequence of the chick embryo, Kuo made the following observations:

• Because it is a bird developing in an egg, the head develops resting on the heart, a function of its phylogenetic standing as a bird
• When the heart begins to beat, as early as 13 hours into development, the head resting on the heart is forced to move up and down passively in response to the expansion and contraction of the beating heart; these head movements appear on days 3 and 4 of development;
• Food getting activities, e.g., opening and closing of the bill, thrusting and clapping, etc., occur on the sixth or seventh developmental day. Of course, these are not actually pecking movements, since there is nothing present to peck at: “But who can deny that the opening and closing of the bill, bill thrusting and lifting and bending of the head which appear as early as the third or fourth day of incubation are primordial movements which after hatching become component parts of pecking reactions?” (1932d, p. 113)
• Beak opening and closing is accompanied by swallowing of surrounding fluids.

To summarize, Kuo observed that long before the very life of a baby chick depends on its being able to lower and raise its head (in response to a grain of food), open and close its beak, and swallow, in a coordinated fashion – i.e., engage in food-getting behavior – the baby chicken has already done those very same things. We fully agree with Kuo’s assessment that, “The data ... cannot fail to show the tremendous influence of prenatal development upon postnatal behavior” (1932d, p. 120). It is important to underscore that this statement was based strictly on observation, not on mere experimental manipulation.

Of course, one may question why the chicken initially pecks at the grains. T. C. Schneirla, a later proponent of Kuo's approach, provided a means of addressing this question in terms of his “approach-withdrawal hypothesis (AW)” (1959), which suggests that early in their lives, newly born animals make approach responses to weak stimulus sources and withdrawal responses to intense stimulus sources. To a baby chick, a kernel of grain is a small or weak visual stimulus (subtending a small retinal area); the chick thus approaches it, but lacking arms and hands can only explore it with its beak. It tastes good and the chicken is reinforced for subsequent grain pecking. The chick is also attracted visually to its own fecal matter, also a small round visual stimulus. These pecks are accompanied by obvious distaste reactions – the fecal matter is an intense chemical package – and the chick subsequently withdraws from its own feces.

This analysis is lent weight by an experiment (Wallman, 1979) in which the feet of newly hatched chicks were fitted with booties, which prevented them from seeing their own toes and claws – also small visual stimuli. These animals were poorer at pecking small mealworms than were undisturbed chicks. Apparently, experience with naturally occurring small visual stimuli – the chicks’ own toes – is a prerequisite for successful food getting. It is worth noting that in his later writing, Kuo acknowledged the significance of Schneirla’s AW formulation (Kuo, 1967, p. 125).

Is food getting by chickens an instinct? Unless one conducts observations such as these we are forced to conclude so. This is why Kuo referred to instinct psychology as a “lazy” science and why he said his work began where the instinct psychologists’ ended. Rather than examine newly hatched chicks – the genetic method of his day – to determine the origins of pecking behavior: “The real nature of behavior cannot be understood unless its underlying physiology and the entire developmental history are known” (1932d, p. 120, emphasis added). Who today would deny the truth of this statement? What is often neglected, even by contemporary researchers, is that self-stimulation – such as that which occurs in the developing chicken or duck embryo – is an important factor in the establishment of behavior patterns (Gottlieb & Kuo, 1965; see also Gottlieb’s contribution in this issue, regarding the background of self-stimulation). Kuo’s work does not demonstrate that behaviors are learned in the egg. Rather, these activities in the egg are “non-obvious experiential precursors” of later behavior.

(Excerpted from Greenberg, G. & Partridge, G. (2000). Prolegomena to Praxiology Redux: The Psychology of Zing-Yang Kuo. From Past to Future: Clark Papers on the History of Psychology. Vol 2(2). From Instinct to Epigenesis: Lessons from Zing-Yang Kuo (pp. 13-37).

• The Magazine
• Stay Curious
• The Sciences
• Environment
• Planet Earth

Schrödinger’s Cat Experiment and the Conundrum That Rules Modern Physics

Why schrödinger (figuratively speaking) put his cat in the box — and why it may never get out..

Long before cats conquered the internet, two of the greatest physicists of our time  — Erwin Schrödinger and Albert Einstein — devised what almost seems like an evil thought experiment.

It goes something like this: You have a cat in a completely sealed box impervious to any observation from outside. Inside is a kind of device involving a Geiger counter, poison, and radioactive material whose atoms may or may not enter a state of decay in equal probability over the course of an hour. If one atom does decay, the Geiger counter detects the radiation and triggers a hammer that breaks open the vial of poison, killing the cat. If no atom decays, then the cat lives.

Of course, the device was only theoretical. Schrödinger developed the scenario in a discussion with Einstein in response to misinterpretations of quantum mechanics at the time. It was a way to describe how a concept that seemed to apply to minute electrons in atoms might apply to a complex object in the macroscopic world — in this case, a cat.

While Schrödinger’s cat remains something of an infamous thought experiment, the original equation he originally derived the scenario from has gone on to represent the foundation of quantum mechanics. It involves the idea that something can be in two simultaneous states and only becomes one or the other when observed, detected, or even when it interacts with other particles. That fundamental theory of physics has modern-day applications that include everything from supercomputers to chemistry and superconducting magnets. 

“[The Schrödinger equation] is like a modern version of Newton’s law,” says Chen Wang , an assistant professor of physics at the University of Massachusetts Amherst.

The Theory in Question

In the 1920s, Schrödinger and other physicists were concerned with a problem that couldn’t be explained by classic physics. The smaller a particle becomes, the less clear its position or speed becomes.

“Quantum mechanics adds a level of fuzziness to the position of particles,” says Wang.

Electrons form the base of the theory — in particular, the single electron in a hydrogen atom. Whereas scientists previously described electrons as orbiting the nucleus in an atom, quantum physicists observed that things weren’t quite this simple. Rather, they seemed to exist in several places at once. Or they blinked back and forth, between specific areas without appearing in between. In fact, the only thing you can say for sure is that an electron is not in one place at one time.

“It’s fundamentally uncertain where exactly the position [of an electron] is,” Wang says. 

Instead, you have to describe an electron’s position as a wave function, or a probability distribution that describes where the electron is more likely to occur. The term superposition in quantum physics is used to describe how an electron in this case can seem to exist in multiple positions at the same time.

Tying Things Up

If you’re not lost yet, the idea gets even wilder when you add an extra electron. In helium, for example — which has two electrons — each one can only be described as probably being in a given area at a given time. But they also can interact and affect each other despite their distance in a process known as quantum entanglement, or “spooky action at a distance,” as Einstein called it.

Another way to think of it is that changing the state of one electron means the state of the other has to change as well.

“The description for two electrons cannot be directly produced by thinking of two independent shapes,” says Frederick Strauch , a physicist at Williams College in Massachusetts. “We can think of them as somehow jumping between the different shapes.”

Nine Lives or Two States?

The scenario involving Schrödinger’s dead — or undead — cat in a box involves a thought experiment to describe how the state of electrons might conceivably affect something much larger, in the macro world. He created it in response to a theory of quantum mechanics by other physicists called the Copenhagen interpretation to show the potential shortcomings of their view. 

Since we can’t see in the box at the end of the hour or send any type of probe inside, according to the Copenhagen theorists, the radioactive atoms remain in a superposition of both decay or non-decay. The cat, in turn, depends on this superposition, as we don’t know whether it’s alive or dead. In a quantum sense, its superposition remains in both states at the same time, as a wave function that is both alive and dead. The entanglement is represented by the connection between the radioactive atoms and the cat, says Strauch.

When we open the box and look inside, or if the outside world somehow interacts with the inside of the box, the wave function is forced to collapse into one state, and the cat becomes dead or alive.

The thing is, Schrödinger didn’t actually mean for the situation to be taken seriously. The fact that an observing scientist’s curiosity could kill the cat was meant to show how an earlier interpretation of quantum mechanics was ludicrous.

“He’s kind of alluding to the fact that the theory is sort of silly to apply to the macroscopic world,” Wang says. “There might be something missing.”

But his thought experiment has since taken on a life (or death) of its own, with many people believing that the cat would be dead and alive at the same time. The only shortcoming of such a thought experiment may rest in our technical inability to carry such an experiment out.

Cats, Present, and Future

Even if Schrödinger himself didn’t believe the cat theory was possible, modern researchers are trying to put some of these theories into practice. In 2016, Wang and his colleagues managed to demonstrate that it’s possible to entangle multiple particles. They managed to measure the entanglement of up to 80 photons, or light particles, placed in special boxes connected by a supercurrent that flows without voltage. In basic terms, it meant the spin they put on photons in one box could be observed in the other box even though they hadn’t spun the latter. Photons without spin also existed in both boxes. Metaphorically, it’s like a live cat and a dead cat were found in two different boxes that were correlated.

Quantum mechanics is already leading to practical applications. Quantum computing is one method, in which the harnessing of superposition and entanglement allows faster calculations than classical computers. Strauch says there are many potential applications of this, but researchers are already on the cusp of using them to calculate chemical formulas in a virtual space to design drugs.

But it may still take a long time before researchers figure out a way to run Schrödinger's experiment. If they ever do, and even the man himself thought this was unlikely, then it could show how the microscopic quantum world might affect the macroscopic world.  

The Quantum Internet Will Blow Your Mind. Here’s What It Will Look Like

Why Quantum Mechanics Still Stumps Physicists New Theory of Everything Unites Quantum Mechanics with Relativity ... and Much More

Already a subscriber?

Register or Log In

Keep reading for as low as $1.99!

Sign up for our weekly science updates.

Save up to 40% off the cover price when you subscribe to Discover magazine.

What did Schrodinger's Cat experiment prove?

Category: Physics      Published: July 30, 2013     Updated: November 27, 2023

By: Christopher S. Baird, author of The Top 50 Science Questions with Surprising Answers and Associate Professor of Physics at West Texas A&M University

"Schrodinger's Cat" was not a real experiment and therefore did not scientifically prove anything. Schrodinger's Cat is not even part of any scientific theory. Schrodinger's Cat was simply a teaching tool that Schrodinger used to illustrate how some people were misinterpreting quantum theory. Schrodinger constructed his imaginary experiment with the cat to demonstrate that simple misinterpretations of quantum theory can lead to absurd results which do not match the real world. Unfortunately, many popularizers of science in our day have embraced the absurdity of Schrodinger's Cat and claim that this is how the world really works.

In quantum theory, quantum particles can exist in a superposition of states at the same time and collapse down to a single state upon interaction with other particles. Some scientists at the time that quantum theory was being developed (1930's) drifted from science into the realm of philosophy, and stated that quantum particles only collapse to a single state when viewed by a conscious observer. Schrodinger found this concept absurd and devised his thought experiment to make plain the absurd yet logical outcome of such claims.

In Schrodinger's imaginary experiment, you place a cat in a box with a tiny bit of radioactive substance. When the radioactive substance decays, it triggers a Geiger counter which causes a poison or explosion to be released that kills the cat. Now, the decay of the radioactive substance is governed by the laws of quantum mechanics. This means that the atom starts in a combined state of "going to decay" and "not going to decay". If we apply the observer-driven idea to this case, there is no conscious observer present (everything is in a sealed box), so the whole system stays as a combination of the two possibilities. The cat ends up both dead and alive at the same time. Because the existence of a cat that is both dead and alive at the same time is absurd and does not happen in the real world, this thought experiment shows that wavefunction collapses are not just driven by conscious observers.

Einstein saw the same problem with the observer-driven idea and congratulated Schrodinger for his clever illustration, saying, "this interpretation is, however, refuted, most elegantly by your system of radioactive atom + Geiger counter + amplifier + charge of gun powder + cat in a box, in which the psi-function of the system contains the cat both alive and blown to bits. Is the state of the cat to be created only when a physicist investigates the situation at some definite time?"

Since that time, there has been ample evidence that wavefunction collapse is not driven by conscious observers alone. In fact, every interaction a quantum particle makes can collapse its state. Careful analysis reveals that the Schrodinger Cat "experiment" would play out in the real world as follows: as soon as the radioactive atom interacts with the Geiger counter, it collapses from its non-decayed/decayed state into one definite state. The Geiger counter gets definitely triggered and the cat gets definitely killed. Or the Geiger counter gets definitely not triggered and the cat is definitely alive. But both don't happen.

Roger Penrose, a Nobel Prize winner and one of the most brilliant physicists of the last sixty years, wrote about Schrodinger's Cat in his book The Road to Reality as follows: "So the cat is both dead and alive at the same time! Of course such a situation is an absurdity for the behavior of a cat-sized object in the actual physical world as we experience it... There is a 50% chance that the cat will be [definitely] killed and a 50% chance that it will [definitely] remain alive. This is the physically correct answer, where 'physically' refers to the behavior of the world that we actually experience." Penrose goes on to explain that any physical theory or philosophical interpretation of quantum physics that leads to the cat being both dead and alive at the same time must be a faulty theory or interpretation, because that is not what happens in the real world.

Despite that fact that the Schrodinger's Cat story is not a real experiment, does not prove anything, does not match physical reality, and was intentionally designed to be absurd, this line of thinking does indeed lead to a meaningful question. Why does the cat in this setup not end up in a state of dead and alive at the same time in the real world? In other words, why does a measurement collapse a quantum object from a superposition of states to a single definite state? This question has not yet been fully answered by quantum physics. This is known as the "measurement problem" of quantum physics. Note that the measurement of a quantum object in a superposition of states collapsing down the object to a definite single state is very well predicted by the mathematics of quantum physics. Therefore, the "measurement problem" is more a problem of philosophical interpretation and incomplete scientific explanation, than a problem of the theory being incorrect.

In summary, quantum state collapse is not driven just by conscious observers. Unfortunately, many popular science writers in our day continue to propagate the misconception that a quantum state (and therefore reality itself) is determined by conscious observers. They use this erroneous claim as a springboard into unsubstantial and non-scientific discussions about the nature of reality, consciousness, and even Eastern mysticism. To them, Schrodinger's Cat is not an embarrassing indication that their claims are wrong, but proof that the world is as absurd as they claim. Such authors either misunderstand Schrodinger's Cat, or intentionally misrepresent it to sell books.

Topics: Schrodinger's Cat , observation , quantum , superposition , wavefunction collapse

Feline Friendly? New Psychology Study Shows How to Build Rap-Paw With Your Cat

A team of psychologists at the Universities of Sussex and Portsmouth have purr-fected the art of building a bond with cats.

The new study, ‘The role of cat eye narrowing movements in cat-human communication,’ published online in the  Nature  journal  Scientific Reports , has shown for the first time that it is possible to build rapport with a cat by using an eye narrowing technique with them.

This eye-narrowing action by humans generates something popularly known as a cat smile – the so-called “slow blink” – and seems to make the human more attractive to the cat. Eye narrowing movements in cats have some parallels with the genuine smile in humans (the Duchenne smile), as well as eye narrowing movements given in positive situations in some other species .

The team, led by Dr. Tasmin Humphrey and Professor Karen McComb, animal behavior scientists at the University of Sussex, undertook two experiments.

The first revealed that cats are more likely to slow blink at their owners after their owners have slow blinked at them, compared to when they don’t interact at all.

The second experiment, this time with a researcher from the psychology team, rather than the owner, found that the cats were more likely to approach the experimenter’s outstretched hand after they’d slow blinked at the cat, compared to when they had adopted a neutral expression.

Taken together, the study shows that this slow blinking technique can provide a form of positive communication between cats and humans.

The study found:

• Cats were more likely to slow blink at their owners if their owners had slowed blinked at them, compared to when the owner was present in the room but not delivering a slow blink stimulus.
• Cats were more likely to slow blink when an unfamiliar experimenter slow blinked at them, compared to when they had maintained a neutral expression.
• Cats preferred to approach an experimenter after they had slow blinked at the cat than if they had maintained a neutral expression.

Professor Karen McComb, from the School of Psychology at the University of Sussex, who supervised the work, said: “As someone who has both studied animal behavior and is a cat owner, it’s great to be able to show that cats and humans can communicate in this way. It’s something that many cat owners had already suspected, so it’s exciting to have found evidence for it.

“This study is the first to experimentally investigate the role of slow blinking in cat–human communication. And it is something you can try yourself with your own cat at home, or with cats you meet in the street. It’s a great way of enhancing the bond you have with cats. Try narrowing your eyes at them as you would in a relaxed smile, followed by closing your eyes for a couple of seconds. You’ll find they respond in the same way themselves and you can start a sort of conversation.”

Dr. Tasmin Humphrey, a PhD student in the School of Psychology at the University of Sussex during the work, was the first author of the study. She said: “Understanding positive ways in which cats and humans interact can enhance public understanding of cats, improve feline welfare, and tell us more about the socio-cognitive abilities of this under-studied species.

“Our findings could potentially be used to assess the welfare of cats in a variety of settings, including veterinary practices and shelters.

“In terms of why cats behave in this way, it could be argued that cats developed the slow blink behaviors because humans perceived slow blinking as positive. Cats may have learned that humans reward them for responding to slow blinking. It is also possible that slow blinking in cats began as a way to interrupt an unbroken stare, which is potentially threatening in social interaction.”

Dr. Leanne Proops at University of Portsmouth , who co-supervised the work, said: “It’s definitely not easy to study natural cat behavior so these results provide a rare insight into the world of cat-human communication.”

How the experiments worked

Two experiments were conducted to explore the significance of the slow blink in cat–human communication.

The first experiment included a total of 21 cats from 14 different households. Fourteen different owners participated in experiment 1. Ten of the cats were male and 11 of the cats were female, with cat age ranging from an estimated 0.45–16 years. The experiments took place in each cat’s home. The psychologist advised the cat’s owner on how to slow blink. Once the cat had settled down in one place, the psychologist asked the owner to either sit approximately 1 m away from the cat.

Experiment 2 included a total of 24 additional cats. Twelve cats were male and 12 cats were female, with cat age ranging from an estimated 1–17 years old. The cats included in the final analyses were from eight different households. In this experiment, the researcher, who was unfamiliar to the cat, either slow blinked at the cat or adopted a neutral face without direct eye contact. This experiment also tested which context the cats preferred to approach the unfamiliar experimenter, by them offering the cat a flat hand with palm faced upwards whilst sat or crouched directly opposite the cat. Both experiments were video recorded.

Cat psychology – the existing context

In the new paper, the authors provide some context for their findings. The psychology of cats hasn’t been studied as extensively as dogs, but what is already known includes:

• That cats have been shown to attract and manipulate human attention effectively through ‘solicitation purring’.
• That cats can discriminate their name from other words, even when unfamiliar humans are calling.
• That cats may be sensitive to human emotional cues, and will rub or butt their head against an owner who feels sad.

‘The role of cat eye narrowing movements in cat-human communication’ by Tasmin Humphrey, Leanne Proops, Jemma Forman, Rebecca Spooner and Karen McComb, published in  Scientific Reports,  is open access.

Reference: “The role of cat eye narrowing movements in cat–human communication” by Tasmin Humphrey, Leanne Proops, Jemma Forman, Rebecca Spooner and Karen McComb, 5 October 2020, Scientific Reports . DOI: 10.1038/s41598-020-73426-0

Related Articles

New research analyzes the types of relationships between cats and their owners – take the quiz yourself, brains of binge-drinkers have to work harder to feel empathy for others, pets, touch and covid-19: why our cats, dogs and other furry friends are lifesavers, searching for a more accurate classification of the dimensions of psychopathology, protective factors are important in preventing violence in veterans, email data reveals global migration trends, theoretical model on the evolution of cooperation, 7 million year old footprints reveal elephant social structure from the past, the less birds know, the better.

I’ve known that for nearly 40 years. I noticed it when I kissed at my cats from across the room and they would always give a slow blink back. I call their blink “kittie kisses”. I also do it to ferals I want to trap as to make them feel more comfortable with my presence.

I have a young tomcat, he blinks at me all the time. When I blink back I see his response of awareness. When i call for him, to come and eat he’s around in a couple minutes.

quite often I see “modern scientific experiments” with animal behavior, and I think “are these folks autistic?” They “eye-narrowing” works best when a complete blink is included, within a slow “narrowing” episode. Blinking, but not “narrowing” alone, also works with dogs that you have not met, as they are approaching you, and sometimes as you approach them; however, with a dog you have not met, simply breaking eye contact can also “relax” them – stare them in the eye, and it can be a “contest” or a “threat”. My best friend in high-school bred my Black Lab. He told me cats are dumb: don’t know their name, won’t come when called, etc. I called my cat’s name, and she looked at me. I said “come here” and she did. (note that no, I never did this before, never “called” them for feeding time, etc.) My friend was astonished. Her sister knew when I was sad, and would come and sit in my lap and give me a “bath” (licking me like I was her kitten). My two dogs could understand complex sentences (they spent more time with me, traveling in the car with me everywhere, etc.). It blew my mom and her boyfriend’s mind. I never simply gave them “command words” (except the word “OK” – which really had no specific meaning – it changed with the context of the situation and they always understood the intended meaning); I always talked to them in full COMPLEX sentences. Once, while cooking in the kitchen, my mom told me to put them outside because they were “underfoot”. I asked “If I can get them to stay in the other room while we cook, if that OK?” She looked at me like I was delusional. I never “trained” them to do such a thing – not even remotely like that. But I had learned how to COMMUNICATE with them. I simply said, in spoken English, “go lie down in the living room and wait until we finish cooking in here” while pointing to the living room. They immediately did just that and watched us until we were done. My mom was flabbergasted. Training “pets” is what STUPID people do. They are stupid because they think they are “better” and their “pet” is dumb. Dumb is a lack of capability. Stupid is an inability or unwillingness to understand despite having the capability. When a non-human life form is treated like a FRIEND and rapport is built, mammals and birds can communicate with each other both within a species and across species (recently “modern scientists” “confirmed” that creatures in the “wild” do just that, but as a “wilderness” hiker, that was old news to me), including with humans. It is the majority modern humans that have lost this ability, due to arrogance, I believe. These are highly evolved creatures (cats, dogs, and more). No they don’t have language, don’t understand trigonometry, and don’t have opposing thumbs to help them build log houses with metal nails, but some are still smarter than some of the humans I know (Tuck Frump)

Oh, yea, forgot to add: “pets” are “entertainment slaves”. “friends” are “friends”. (and sorry for the typos in the previous post)

Stupid me. I said “they don’t have language”. This is NOT a fact. HUMANS don’t understand their language. It may be (is likely) far more simple than ours. I can tell something about what a cat wants by the type of “meow” it makes. Many an in-tune parent can also tell what thier baby needs by the “type” of crying it makes. But these “sounds” that correspond to meaning do not nessesarily comprise a “language” in the sense of an “infinite open system with limits” that humans have – a system that can convey nearly any meaning. However, in the case of birds, whales, dolphins, etc., humans have been trying to “understand” the “language of communication” between these creatures, with little to no success. Yet they (the ones held prisoner at Sea World, at least) can understand us when we speak! Whose the good boy there?

good to know.

Save my name, email, and website in this browser for the next time I comment.

Type above and press Enter to search. Press Esc to cancel.