Robert W. Doty - US grants

By placing its hand upon a plate the highly trained macaque indicates its readiness to attempt detection of a signal which will occcur at random within 10-20 sec. If it removes its hand from the plate within 150-1000 msec after signal onset, it is presumed to have detected the signal and is rewarded. Random removals are punished with air puff and delay of next available attempt. Animals trained in this manner become highly reliable psychophysical observers and "report" in this fashion thresholds which remain constant over a period of many months for electrical excitation of striate cortex via permanently implanted electrodes. Against this background they are then tested for threshold to stimulation, 0.2-msec pulses, 50-100 Hz, with a microelectrode passed sequentially through the cortical laminae. Laminar position is monitored by background activity and photically evoked response. Two minima in threshold usually occur in such traverses. They are usually 2-3 fold less than at adjacent sites and are presently thought to be associated with the inner and outer bands of Baillarger. Transition from nondetection to detection can occur for stimulus changes less than or equal to 1.0 MuA. Reports from human subjects (Brindley and Lewin; Dobelle and Mladejovsky) emphasize the uniformity of the effect produced by electrical excitation any place in striate cortex, and data from macaques are concordant with this. Thus, an unusual opportunity is at hand: a sharp threshold transition consequent to a precisely controlled physical input at a fixed and known location in the nervous system, producing an immediately assessable behavioral consequence. By assaying the synaptic events with current source density analysis in the vicinity of the stimulating microelectrode, as well as the single unit activity engendered by detectable versus nondetectable pulses, it should be possible to arrive at a firm definition of the neural events initiating a conscious perception. Additionally, data will be acquired concerning interlaminar transactions and the neural response to interjected activity, facts of relevance to any prosthetic or therapeutic use of electrical stimulation of the central nervous system, and for understanding the organization of visual processes.

The proposed experiments incorporte two heretofore separate but highly productive lines of attack on defining and elucidating the neural substrate of memory, a problem of vast import to fields as diverse as education, psychiatry, neurology and gerontology. The experiments utilize, on the one hand, the sophisticated techniques recently developed for testing memory in nonhuman primates. These have demonstrated a striking similarity between macque and man in the processes of visual memory, both in its normal operation and in the nature of the impairment consequent to homologous cerebral loss. On the other hand, these experiments exploit the extraordinary possibilities afforded by the "split-brain" approach, whereby information can be supplied to or demanded from each cerebral hemisphere individually, thus allowing comparison of performance by "intact" versus damaged hemispheres in the same animal, the tracing and comparison of different pathways for interhemispheric communication, etc. Macaques with surgically transected optic chiasm view with one or the other eye, or each seriatim, a number of pictures which they are subsequently called upon to identify when viewed by the same or other eye and hemisphere. Interhemispheric communication is limited by cutting one of the forebrain commissures. Lesions are to be created reversibly by cooling, local anesthetization, or electrical tetanization; and irreversibly by subpial suction or transection. The primary goals are to study 1) how initial distribution of visual mnemonic input to one or the other hemisphere affects the retrieval of these memories when achieved ipsilaterally, contralaterally or bilaterally, comparing interhemishpheric communication via anterior commissure versus splenium of the corpus callosum in this regard; and 2) to define how unilateral, reversible or irreversible, elimination of various cerebral structures having a suspected role in visual memory will alter the mnemonic capability of that hemisphere, particulary as to whether "recording" versus "retrieval" is differentially affected. By such means, it should be possible to define the structures essential to visual mnemonic processing, and to some degree the nature of their role therein.

Much contemporary work focuses on the hippocampus as a structure of unique importance for memory. It is thus of obvious interest to understand what information passes from one hemisphere to the other by way of the hippocampal commissure (HC). Examination of this question, however, has heretofore been deemed unapproachable because of the close adherence of the dorsal HC (dHC) to the overlying corpus callosum (CC) and insufficient knowledge about the connectivities of the fibers that constitute the dHC and contiguous parts of the CC. This impracticality is now relieved by emerging understanding of the relevant interhemispheric pathways, which suggest a new surgical strategy, devised and tested in this laboratory on macaques, that will permit testing interhemispheric transfer of visual/mnemonic information across the dHC in isolation, or in conjunction with the splenium of the CC. The dHC-splenium combination has already been extensively studied in this regard; however, it is entirely unknown whether the dHC shares, duplicates, or augments the capabilities of the splenium or, indeed, whether it accounts for some of the properties previously attributed to the latter. The transfer to one hemisphere of visual discriminations learned by the other and interhemispheric recognition of previously presented visual images will be tested; and, in the same animals, the complex origin and destination of fibers that constitute the dHC, splenium, and anterior commissure-all of which have components arising in common from the parahippocampal gyrus-will be examined. The results should advance understanding of interhemispheric processes and their relation to memory. Moreover, since interhemispheric processing has been shown to be abnormal in schizophrenia, the results will contribute fundamental information pertinent to understanding of some of the symptoms of this complex mental disorder.

A massive body of evidence, from neuropathology, studies of "split- brain" patients, of "divided visual field" studies in normal subjects, and now from neuroimaging, is all consistent in showing that in right- handed individuals the primary neuronal processing of verbal material takes place in the left hemisphere, while analysis of visual form often proceeds initially in the right hemisphere This situation of "hemispheric specialization" most likely accrues from the Act that communication between the hemispheres is constrained by the very small size of most of the nerve fibers in the forebrain commissures, making it much quicker to process information intrahemispherically than to shuttle back and forth between the hemispheres. Concentrating processing in one hemisphere must also mean that the neuronal facilitation of attention is "focussed" in the relevant hemisphere and that, as is well-known for attending to visual space, disengaging this attentive focus from one locus and concentrating it on another takes a measurable amount of time. Surprisingly, this temporal cost of switching attention between the cerebral hemispheres has never been measured, and it is the intent of this proposal to do so. We will use memory for words and for nonobjective colored images to tap each hemisphere's analytic and hence attentive predilection, already having shown the high degree of similarity in the mnemonic processing of these two disparate types of material. Any increment in reaction time consequent to switching the demands of memory from one to the other type of material can be detected. This will be done using a divided visual field paradigm in which the words or images are directed initially to one or the other hemisphere, and the "switching time" can thus be disentangled from other aspects of interhemispheric communication. Knowing this switching time and its properties should greatly advance understanding of cognitive events that draw upon or demand jointly the specialized skills of the two hemispheres; and it seems likely that the ability to define this time would have wide application in evaluating a variety of cognitive disturbances, including schizophrenia and dyslexia.