Interhemispheric Transfer and the Completion of Reversible Operations in Non-Conserving Children
Rhawn Gabriel Joseph, Ph.D. and R. E. Gallagher, Ph.D.
USH/The Chicago Medical School


Evidence that concerns interhemispheric transfer deficits in young children is reviewed, and it is suggested that the late maturation of the corpus callosum may prevent the transfer and thus verbal expression of knowledge concern- ing equivalence relationships in non-conservers, operations normally performed by the right hemisphere. Because previous findings with 'split-brain' patients indicate that interhemispheric transfer can be achieved when questions are phrased in the context of a reward (e.g., 'preference'), questions concerning the conservation of a non-equivalent volume were posed to 65 non-conservers in terms of an imagined reward (preference condition). Questions also were phrased so as to require comparative-relational determinations as well as expectation of an empirical reversal of the stimulus condi tions so as to determine differences/similarities in the reasoning process. Significantly more non-conservers demonstrated an awareness of empirical reversal or performed the preference relational operation correctly as compared to conditions that required comparative-relational judgments. The evidence also suggests the presence of two fundamentally distinct cognitive approaches to the problem, which appear to resist integration in children of this age group.


Conservation of volume is the ability to realize that two different containers may have the same volume, even though the containers are of a different shape.

Piaget demonstrated that if you show a child two beakers of water, one of which is short and fat, the other tall and thin, and ask the child which beaker contains more water, children under the age of 7 will generally choose or say 'the tall one', even though both contain the same amount of water.

According to Piaget, prior to the development of the understanding of conservation, the thought of the child is guided by "appearances" rather than logical thought or deductive reasoning. That is, appearances (or perception) dominates, such that, for example, if an object such as a ball of malleable clay is stretched out and elongated, the child may believe that the elongated clay is greater in size and has more volume than the ball of clay.

Piaget argued that children fail to "conserve" and are fooled by the differences in the size of the container, because children younger than 7 have not developed the cognitive and symbolic ability to conserve volume or to engage in reversible operations (to think back and make mental comparisons).

In order to engage in conservation, the child, therefore, must become capable of engaging in though patterns that allow for "reverse" thinking; i.e. to remember that the volumes of water were the same when in the same size contains, and that the ball of clay contains the same amount of clay despite changed in its shape.

Once a child reaches the age of 7 or 8, they become increasingly capable of performing reversible operations. Thus, if you pour water from the short wide glass into the tall, thin glass, the child realize there is the same amount of water as before, despite the illusion of appearances.

However, the nature of Piaget's conservations tasks almost always requires that decision making take place within the domain of language; the child is "asked" to engage in "verbal reasoning" and to indicate his or her understanding through language; that is, what is examined is verbal thought, and only those reasoning processes that can be communicated through language. Perception and reasoning, however, can also take place through the non-verbal domain (Joseph, 1982) and depending on the task, this may require the exchange of information between the right and left hemisphere; the left being dominant for language.

In several recent theoretical and experimental reports it has been suggested that human psychic development and/or the expression of certain cognitive abilities parallels cerebral development and the myelination of the corpus callosum (Epstein, 1974; Gallin, 1977; Gallagher & Joseph, 1982; Joseph, 1982; Joseph, Gallagher, Holloway, & Kahn, 1984; Kraft, Mitchell, Languis, & Wheatley, 1980; Lecours, 1975; Meyersburgh & Post, 1979; Milner, 1967). For example, Epstein (1974) and Meyersburgh and Post (1979) have argued that neural fiber pathway development and gross increments in brain tissue volume may be directly related to learning ability and the stages of cognitive development as defined by Piaget. In addition, Joseph et al. (1984) have provided evidence that indicates that information processing in certain modalities may be deficient in young children due to immaturity of the callosum as well as the inferior parietal lobule--a brain area intimately involved in cross-modal (auditory, visual, somesthetic) assimilation (Joseph, 1982). In consequence, information processing and the verbal expression and description of operations performed in other brain regions are often incomplete and/or erroneous due to incomplete transfer (e.g., from the right to left hemisphere). There is also evidence that strongly suggests that the immature corpus callosum pathway that takes over 10 years to mature--may be partially responsible not only for deficits in the interhemispheric transfer of learning, but also the failure of pre-operational children to demonstrate knowledge of operations involved in the conservation of liquid volume (Finlayson, 1976; Kraft et al., 1980).

There are now a number of research reports that indicate that young children function as though their right and left cerebral hemisphere were not fully interconnected, such that they appear to have partially split-brains (Finlayson, 1976; Galin, Diamond, & Herron, 1977; Galin, Johnstone, Nakell, & Herron, 1979; Gallagher & Joseph, 1982; Joseph et al., 1984; Joseph, Lesevre, & Dreyfus-Brisac, 1976; Kraft et al., 1980; Salamy, 1978). In this regard, the left hemisphere of a young child does not always seem to know what is going on in the right (Joseph et al., 1984). For example, Galin et al. (1977, 1979) have shown that 3-year-olds are deficient in cross-modal textural identification and tactile identification when interhemispheric transfer was required, whereas Joseph et al. (1984) found that the ability to transfer information between the right and left brain increases with age. Moreover, children age 4 (as compared to those 7 or 10) performed deficiently when their left hemisphere was required to verbally describe pictures that had been transmitted tachistoscopically to the right such that their responses were contaminated by confabulatory ideation. In addition, Joseph et al. (1976) and Salamy (1978) demonstrated electrophysiologically that interhemispheric linkage in infants and transfer in children below age 3 is incomplete or non-existant in the specific modalities assessed.

Conservation, Confabulation and Interhemispheric Transfer

Given the findings briefly reviewed above and the histological evidence that demonstrates that the corpus callosum (which inter-connects the two brains) is not com-pletely matured or myelinated until the end of the first decade, it appears that although some information channels are intact, children function is some respects as though they have 'split-brains.' If, indeed, the left hemisphere is unable to give an accurate account of operations performed by the right until the callosum is sufficiently matured, it is likely that higher-order cognitive abilities mediated by the right actually may appear earlier than is suspected.

There is some evidence that suggests that due possibly to callosal immaturity children may be inappropriately engaging or relying on the language dominant left hemisphere so as to provide verbal reports and thus are unable to give accurate descriptions of what is known or what has been performed by the right cerebrum (Gallagher & Joseph, 1982; Joseph et al., 1984). Indeed, Kraft et al. (1980), who recorded EEGs from the right and left parietal area of children 6 to 8, found during the performance of a Piagetian conservation task that those who were able successfully to perform conservation operations responded with greater right hemisphere involvement during the response phase than those who could not, whereas Gallagher and Joseph (1982) found that non-conserving children engage the left cerebral hemisphere more than the right and more so than adults.

When coupled with the data presented by Gallagher and Joseph (1982) and colleagues (Joseph et al., 1984), it appears that young children fail to demonstrate successfully these operations because the left hemisphere, which is required to provide a response in the absence of right hemisphere input, confabulates an explanation based on current and externally available information. This strengthens the supposition that young children may be able to adequately process and perform certain operations, but are unable to give accurate verbal descriptions of their (the right hemisphere's) conclusions.

The purpose of the present study was to assess for the presence of correct knowledge concerning equivalence relationships and the ability to perform internal reversals, in pre-operational-nonconserving children. As suggested by Geschwind (1965) and as demonstrated by Gazzaniga and LeDoux (1978), interhemispheric information transfer is facilitated (although the hemispheres are 'split') when questions and/or stimuli are couched in an emotionally arousing ('preference,' Gazzaniga & LeDoux, 1978) or reinforcing context (Geschwind, 1965). Hence, for the purposes of this study, the assessment of conservation ability employed a preference measure as well as the more traditional assessment techniques, as presumably intact callosal fibers, the anterior commisure, or other subcortical pathways may be activated through such a manipulation (cf. Gazzaniga & LeDoux, 1978; Geschwind, 1965; Joseph et al., 1984).



Sixty-six children 4 and 5 years of age who attend day care centers or nursery schools in the vicinity of Chicago, Illinois, participated in the study. They were assigned to one of three groups, with all groups approximately balanced for number and sex.


The testing room contained two chairs and a table low enough to accommodate the child comfortably. Materials used for testing conservation included a small flask that contained colored water, two 6.5 x 5 cm cylindrical glass beakers, and one 19 x 2.5 cm glass cylinder. There were no other distracting stimuli.


Immediately prior to the experiment all participants were given a preliminary test to determine whether they were conservers. Each child was brought individually to the testing room and was seated at the table facing the experimenter. The child was asked to take the flask and pour equal amounts of colored water into the two short glasses. In all cases the experimenter would, if necessary, ensure equivalence by adding an appropriate amount. The experimenter then asked the child to take one of the beakers and pour its contents into the tall beaker. The child then was asked, 'Do these glasses both have the same amount, or does one have more?' One child responded correctly and was excused from further testing. Those who indicate that the taller beaker had more water were classified as non-conservers and then were tested according to one of the three conditions described below.

Condition 1. The experimenter asked the child to pour the liquid from the tall glass back into the short beaker used during the pretest. The child was asked whether the two now held the same amount or whether one had more. When the child confirmed that the two quantities were the same, he or she was asked to pour a little liquid from one short beaker into the other, so that one obviously would contain more. The child then was instructed to pour one of the short glasses (indicating the one with less liquid in it) into the tall beaker and was asked a relational question, 'which glass has more?' The experimenter then asked a question that required an empirical reversal of the presently perceived context, 'Suppose we were to pour it back into the other glass again. Which of the two (short glasses) would have more?' After responding, the child was asked actually to perform this reversal and to compare the quantities. He then was told to return the smaller amount of liquid to the tall cylinder, and the experimenter asked the child to imagine the stimulus in a reinforcing context and to indicate a preference: 'Let's pretend that you are very thirsty and that this is your favorite drink. Which one would you rather have?' These final three questions, relational, reversal, and preferences, were recorded as correct or incorrect as determined by the child's selection of the vessel that did indeed contain the greater volume of liquid.

Condition 2. Procedures for Condition 2 were exactly the same as for Condition 1 except that the reversal situation was eliminated. Thus the relational question was followed immediately by the preference question.

Condition 3. Condition 3 was identical to 1 except that the order of the first and third questions was reversed, so that the preference question was asked first and the one that required a specific relational judgment was asked last.


The results were quite clear and supported the hypothesis. Chi square analyses were applied to the data, with correction factors computed when necessary. The preference question was answered correctly with significantly greater frequency than was the relational question, x2 = 9.98, p < .01. The reversal question also was answered correctly significantly more often than the relational question, x2 = 32.71, p < .005, and the preference question, x2 = 7.87, p < .025. No significant sex differences were found.

Table 1

Correct Preference, Relational and Reversal Responses of Subjects In Conditions 1-3

Condition ---------1-----2-----3



The results demonstrate that non-conserving children maintain an awareness of equivalence relationships, but are more likely to conserve when the situation involves imagining a rewarding experience, as opposed to merely verbally describing a currently perceived static relation between substances. Coupled with the data provided by Gallagher and Joseph (1982) and Kraft et al. (1980), this evidence is consistent with the hypothesis of inappropriate left hemisphere processing, which leads to a performance deficit that only can be overcome when responses are emotionally aroused and, thus, transferred and expressed. In this regard, we believe that presenting the problem in the context of a reinforcer allowed for right hemisphere activation and, thus, inter-hemisphere transfer of this knowledge through pathways that largely may circumvent, but that may include, the immature callosal fibers that connect the two cerebral hemispheres. A similar mode of inter-hemispheric information transfer in a split brain patient has been presented by Joseph et al., (1984).

As noted, a significant number of non-conserving children were capable of performing the empirical reversal and preference inversion problems. However, by directly linking the preference and relational questions to the presently perceived (albeit transformed) stimuli and requiring the children to address their responses accordingly, the possibility that these same children were basing their responses on knowledge that involved empirical return rather than a non-linguistic knowledge of inversion was controlled for. Although both reversal and preference were demonstrated significantly more often than knowledge of relational non-equivalence, the failure of many subjects to respond correctly on both question types mitigates any suggestion that these two processes also may be related in some way. Although both involved communication of knowledge that concerns conservation, the preference response was tied directly to an immediate perception of the stimulus array, whereas the reversal involved responding to the experimenter's questions with regard to a previously perceived or expected stimulus context. Hence, we are presented with two fundamentally different types of knowledge, both of which appear separately yet exist simultaneously in a form that appears to resist integration in children at this age level.

This evidence is consistent with the view that young children are functioning as though they had split brains and suggests that non-conserving children fail to express knowledge associated with the right hemisphere because they appear to be relying on a left hemisphere 'cognitive' strategy due to the linguistic referents traditionally involved with the task. Hence, when understood in terms of functional asymmetry and interhemispheric disconnection, the statement that children do not comprehend equivalence relationship loses its meaning.


Copyright: 1996, 2000, 2010, 2018 - Rhawn Joseph, Ph.D.