DIFFERENTIAL INTERHEMISPHERIC TRANSFER
FOR ABSTRACT AND SPECIFIC VISUAL-FORM INFORMATION

Christopher D. Nicholas & Chad J. Marsolek
University of Arizona

(Presented at the 3rd Annual Meeting of the Cognitive Neuroscience Society, San Fransisco, CA, April 1996)

Introduction

Evidence from long-term memory experiments suggests that relatively independent visual-form subsystems operate in the brain:

An abstract visual-form (AVF) subsystem underlies recognition of abstract categories of form (e.g., A/a vs. S/s) and operates more effectively in the left hemisphere (LH) than in the right (RH).
A specific visual-form (SVF) subsystem underlies recognition of specific instances of forms (e.g., A vs. a) and operates more effectively in the RH than in the LH.

(Marsolek, 1995; Marsolek, Kosslyn, & Squire, 1992; Marsolek, Schacter, & Nicholas, in press)

In this experiment, we test whether similar results will be obtained in short-term memory comparison tasks. More important, we examine whether computational properties of these subsystems determine interhemispheric communication of visual information.

Experiment 1

Two Comparison Tasks:

Abstract Category (AC) comparison task: Decide whether two letters correspond to the same letter category (e.g., A and a -> "same"; A and f -> "different")
Specific Instance (SI) comparison task: Decide whether two letters are physically the same (e.g., A and A -> "same"; A and a -> "different")

Example Displays

Method

Each trial consisted of a stimulus array with 4 characters (see above):

2 letters (e.g., A and a)
2 distractors (i.e., # and #)

The stimulus displays were of two types:

Letters in different visual-fields (AH trials)
Letters in the same visual-field (WH trials)

Letters within a pair varied in their visual similarity:

A and a have relatively dissimilar letter-case structures
S and s have relatively similar letter-case structures

48 subjects performed 256 trials of each comparison task (see above) per experiment:

AC comparison task in one block
SI comparison task in another block

Dependent variables:

d' score and mean response time for correct responses

Predictions

If SVF and AVF subsystems underlie comparisons in the SI and AC tasks, respectively, then:

RH advantage for the SI task in WH trials
No RH advantage for the AC task in WH trials

If a single, undifferentiated subsystem underlies comparisons in the SI and AC tasks, then:

No hemisphere advantages for either task in WH trials

Interhemispheric Communication

When subjects compare two letters, interhemispheric transfer of visual information must take place when the letters are displayed in different visual fields (i.e., across-hemispheres [AH]), but not when they are displayed in the same field (i.e., within-hemisphere [WH]).

AH processing requires interhemispheric transfer	WH processing doesn't require interhemispheric transfer

When the cerebral hemispheres must communicate with each other, performance involves a trade-off between distribution and degradation (Banich & Nicholas, in press)

AH trials:

Advantage:	Hemispheric distribution of information: Reduced processing load per hemisphere (Dimond & Beumont, 1971)
Disadvantage:	Callosal degradation of information: Information must be transferred (Dimond, Gibson, & Gazzaniga, 1972)

WH trials:

Advantage:	No callosal degradation of information: Information does not need to be transferred
Disadvantage:	No hemispheric distribution of information: increased processing load per hemisphere

FIGURE: RESULTS FROM WH AND AH TRIALS (COLLAPSED)

Complexity Interpretations

Hemispheric distribution may be effective only when independent processes can be insulated from each other (Liederman, 1986; Liederman, Merola, & Hoffman, 1986).

The AC task requires perceptual and phonological or name processing:

For two independent processes, the benefits of distribution outweigh the costs of degradation

The SI task requires only perceptual processing:

For a single process, the costs of degradation outweigh the benefits of distribution

Thus, an AH advantage is found for the AC task because it requires two stages of processing, whereas a WH advantage is found for the SI task because it requires only one stage of processing (Banich & Belger, 1990).

Subsystems Interpretation

An AVF subsystem is needed in the AC task. This subsystem should process relatively invariant features effectively, treating minor visual differences as noise that should be ignored:

An AVF subsystem may be relatively insensitive to the noise generated from callosal degradation

An SVF subsystem is needed in the SI task. This subsystem should process visually distinctive information effectively, treating visual differences as essential information:

An SVF subsystem may be relatively sensitive to the noise generated from callosal degradation

Thus, an AH advantage is found for the AC task because the costs of degradation do not outweigh the benefits of distribution, whereas a WH advantage is found for the SI task because the costs of degradation do outweigh the benefits of distribution.

Predictions

If the complexity of the comparison tasks (number of processing stages required) determines the efficacy of interhemispheric communication:

Same pattern of results (AH advantage for AC task; WH advantage for the SI task) for both similar- and dissimilar-case letters

If the computational properties of the processing subsystems needed to perform the comparisons tasks determines the efficacy of interhemispheric communication:

For similar-case letters (e.g., S/s):

not

AH advantage for the AC task

If an SVF subsystem is required for the SI task, and noisy versions of S and s cannot be distinguished effectively by this subsystem because these specific instances have very fine-grained distinctive information that may be affected greatly by collosal degradation, then:

WH advantage for the SI task

For Dissimilar-case letters (e.g., A/a):

No AH advantage for the AC task

If an SVF subsystem is required for the SI task, and noisy versions of A and a can be distinguished effectively by this subsystem because these specific instances have relatively coarse distinctive information that may not be affected greatly by collosal degradation, then:

No WH advantage for the SI task

FIGURE: RESULTS FROM WH AND AH TRIALS (SIMILAR-CASE)

FIGURE: RESULTS FROM WH AND AH TRIALS (DISSIMILAR-CASE)

Experiment 2

In Experiment 1, an AH advantage in the AC task and a WH advantage in the SI task were found for one set of letters, but not for another set. Presumably, this is because the former set, but not the latter, was comprised of letters with similar lower- verses uppercase visual structures.

In this experiment, we test whether case similarity is the crucial difference between letter groups by using the dissimilar-case letters used in Experiment 1 and presenting visually similar instances of the same dissimilar-case letter (e.g., and ; making them like the similar-case letters used in Experiment 1).

Method

Each trial consisted of a stimulus array with 4 characters (see above):

2 letters (e.g., and )
2 distractors (i.e., # and #)

The stimulus displays were of two types:

Letters in different visual-fields (AH trials)
Letters in the same visual-field (WH trials)

Only dissimilar-case letters were used, however different instances in the same abstract categories were always different-font, same-case items (e.g., and ), not different-case items (e.g., A and a) as in Experiment 1.

48 subjects performed 256 trials of each comparison task (see above) per experiment:

AC comparison task in one block
SI comparison task in another block

Dependent variables:

d' score and mean response time for correct responses

Predictions

If the different results between dissimilar- and similar-case letters in Experiment 1 were obtained because those two sets of letters are always processed in qualitatively different manners in the brain that result in different patterns of interhemispheric communication:

No AH advantage for the AC task (as in Experiment 1)
No WH advantage for the SI task (as in Experiment 1)

If the different-font (but same-case) instances of dissimilar-case letters (/) are processed like different-case instances of similar-case letters (S/s):

AH advantage for the AC task
WH advantage for the SI task

FIGURE: RESULTS FROM WH AND AH TRIALS (EXPERIMENT 2)

Summary

Hemispheric asymmetries for Abstract and Specific Visual-Form Comparisons:

Right Hemisphere advantage in the Specific Instance Task
No Right Hemisphere advantage in the Abstract Categorization Task

Comparing different-case versions of similar-case letters (S/s):

Across Hemisphere advantage in the Abstract Category Task
Within Hemisphere advantage in the Specific Instance Task

Comparing different-case versions of dissimilar-case letters (A/a):

No Across Hemisphere advantage in the Abstract Category Task
No Within Hemisphere advantage in the Specific Instance Task

Comparing different-font versions of dissimilar-case letters (

Across Hemisphere advantage in the Abstract Category Task
Within Hemisphere advantage in the Specific Instance Task

Conclusions

Relatively independent subsystems underlie abstract and specific visual-form comparisons

Comparisons that require an Abstract Visual-Form Subsystem:

Are relatively insensitive to the effects callosal degradation
Can take advantage of hemispheric distribution, when the comparison items are different instances in the same abstract category that have a large amount of common visual structure

Comparisons that require a Specific Visual-Form Subsystem:

Are relatively sensitive to the effects callosal degradation
Cannot take advantage of hemispheric distribution, when the comparison items different specific instances but contain very fine-grained distinctive information

Implications

These findings suggest that in order to understand interhemispheric communication and interaction, a computational subsystems approach may be more fruitful than approaches that rely on general principles of communication.

References

Banich, M. T., & Belger, A. (1990). Interhemispheric interaction: How do the hemispheres divide and conquer a task? Cortex, 26, 77-94.\

Banich, M. T., & Nicholas, C. D. (in press). Integration of processing between the cerebral hemispheres in word recognition. In Getting it right: In The cognitive neuroscience of right hemisphere language comprehension. Beeman, M, & Chiarello, C. (Eds.) Hillsdale, N. J.: Lawrence Erlbaum Associates.

Dimond, S. J., & Beumont, G. (1971). Use of two cerebral hemispheres to increase brain capacity. Nature, 232, 270-271.

Dimond, S. J., Gibson, A. R., & Gazzaniga, M. S. (1972). Cross-field and within-field integration of visual information. Neuropsychologia, 10, 379-381.

Liederman, J. (1986). Interhemispheric interference during word naming. International Journal of Neuroscience, 30, 43-56.

Liederman, J., Merola, J., & Hoffman, C. (1986). Longitudinal data indicate that interhemispheric independence increases during early adolescence. Developmental Neuropsychology, 2, 183-201.

Marsolek, C. J. (1995). Abstract-visual-form representations in the left cerebral hemisphere. Journal of Experimental Psychology: Human Perception and Performance, 21, 375-386.

Marsolek, C. J., Kosslyn, S. M., & Squire L. R. (1992). Form-specific visual priming in the right cerebral hemisphere. Journal of Experimental Psychology: Learning, Memory, and Cognition, 18, 492-508.

Marsolek, C. J., Schacter, D. L., & Nicholas, C. D. (in press). Form-specific visual priming for new associations in the right cerebral hemisphere. Memory & Cognition.

Acknowledgements

This research was supported by the National Institute of Mental Health Grant MH53959-01. The research was also supported by the McDonnell-Pew Cognitive Neuroscience Center, the Program in Neuroscience at the University of Arizona, and the Arizona Cognitive Science Program.

Please send correspondence to:

nicholas@u.arizona.edu