A "theory of relativity" for cognitive elasticity of time and modality dimensions supporting constant working memory capacity: involvement of harmonics among ultradian clocks?

1. The capacity of working memory (WM) for about 7+/-2 ("the magical number") serially organized simple verbal items may represent a fundamental constant of cognition. Indeed, there is the same capacity for sense of familiarity of a number of recently encountered places, observed in radial maze performance both of lab rats and of humans. 2. Moreover, both species show a peculiar capacity for retaining WM of place over delays. The literature also describes paradoxes of extended time duration in certain human verbal recall tasks. Certain bird species have comparable capacity for delayed recall of about 4 to 8 food caches in a laboratory room. 3. In addition to these paradoxes of the time dimension with WM (still sometimes called "short-term" memory) there are another set of paradoxes of dimensionality for human judgment of magnitudes, noted by Miller in his classic 1956 paper on "the magical number." We are able to reliably refer magnitudes to a rating scale of up to about seven divisions. Remarkably, that finding is largely independent of perceptual modality or even of the extent of a linear interval selected within any given modality. 4. These paradoxes suggest that "the magical number 7+/2" depends on fundamental properties of mammalian brains. 5. This paper theorizes that WM numerosity is conserved as a fundamental constant, by means of elasticity of cognitive dimensionality, including the temporal pace of arrival of significant items of cognitive information. 6. A conjectural neural code for WM item-capacity is proposed here, which extends the hypothetical principle of binding-by-synchrony. The hypothesis is that several coactive frequencies of brain electrical rhythms each mark a WM item. 7. If, indeed, WM does involve a brain wave frequency code (perhaps within the gamma frequency range that has often been suggested with the binding hypothesis) mathematical considerations suggest additional relevance of harmonic relationships. That is, if copresent sinusoids bear harmony-like ratios and are confined within a single octave, then they have fast temporal properties, while avoiding spurious difference rhythms. Therefore, if the present hypothesis is valid, it implies a natural limit on parallel processing of separate items in organismic brains. 8. Similar logic of periodic signals may hold for slower ultradian rhythms, including hypothetical ones that contribute to time-tagging and fresh sense of familiarity of a day's event memories. Similar logic may also hold for spatial periodic functions across brain tissue that, hypothetically, represent cognitive information. Thus, harmonic transitions among temporal and spatial periodic functions are a possible vehicle for the cognitive dimensional elasticity that conserves WM capacity. 9. Supporting roles are proposed of (a) basal ganglia, as a high-capacity cache for traces of recent experience temporarily suspended from active task-relevant processing and (b) of hippocampus as a phase and interval comparator for oscillating signals, whose spatiotemporal dynamics are topologically equivalent to a toroidal grid.

Hypothesized neural dynamics of working memory: several chunks might be marked simultaneously by harmonic frequencies within an octave band of brain waves.

The capacity of working memory (WM) for up to about seven simple items holds true both for humans and other species, and may depend upon a common characteristic of mammalian brains. This paper develops the conjecture that each WM item is represented by a different brain wave frequency. The binding-by-synchrony hypothesis, now being widely investigated, holds that the attributes of a single cognitive element cohere because electroencephalogram (EEG) synchrony temporarily unifies their substrates, which are distributed among different brain regions. However, thought requires keeping active more than one cognitive element, or WM "chunk," at a time. If there is indeed a brain wave frequency code for cognitive item-representations that are copresent within the same volume of neural tissue, the simple mathematical relationships of harmonies could provide a basis for maintaining distinctness and for orderly changes. Thus, a basic aspect of music may provide a model for an essential characteristic of WM. Music is a communicative phenomenon of "intermediate complexity," more highly organized than the firing patterns of individual neurons but simpler than language. If there is a distinct level of neural processing within which the microscopic physiological activity of neurons self-organizes into the macroscopic psychology of the organism, it might require such moderate complexity. Some of the obvious properties of music--orderly mixing and transitions among limited numbers of signal lines-are suggestive of properties that a dynamic neural process might need in order to organize and reorganize WM markers, but there are a number of additional, nonobvious advantageous properties of summating sinusoids in music-like relationships. In particular, harmonies register a stable periodic signal in the briefest possible time. Thus, the regularity of summating sinusoids whose frequencies bear harmony ratios suggests a particular kind of tradeoff between parallel and serial processing. When there are few copresent waves, at EEG frequencies, this sort of parallel coding retains behaviorally meaningful brief periods. A necessary companion hypothesis is that the brain wave frequencies underlying WM are confined to a single octave; that is, the upper and lower bounds of the band are in the ratio of 2:1. This hypothesized restriction, suggested by an empirical property of EEG bands that has been widely reported but rarely commented upon, has the important property of precluding spurious difference rhythms. A restriction to an octave, of "harmonious" frequency-markers for WM items, also seems consistent with a great deal of behavioral data suggesting that WM comprises a rapidly fading trace process in which only up to three or four item-representations are strongly activated simultaneously. There is also an additional, sequential renewal-or-revision process, within which up to another three or four items are being actively refreshed by rehearsal or replaced. Such serial processing may involve a less stringent octave band crowding problem.

A working memory "theory of relativity": elasticity in temporal, spatial, and modality dimensions conserves item capacity in radial maze, verbal tasks, and other cognition.

It is remarkable that working memory (WM) capacity for numbers of items remains modest, at approximately 7+/-2 (the so-called "magical number"), across a wide variety of kinds of material. Indeed, consideration of radial maze studies together with more traditional memory research shows that WM capacity remains fairly constant whether the items are verbal or visuospatial, and that this same capacity is true of other species as of humans. In contrast to their limited numerousness, WM items are extremely flexible in ways that are here brought under the heading of "dimensionality." Therefore, the physical items represented in WM, can vary widely in any quantitative characteristic and in the temporal pace at which they are encountered. Combinatorial considerations suggest that WM numerousness results from evolution of a middle ground between a sterile parsimony and an overwhelming excess, for organizing neurocognitive associations. Such natural selection seems likely to have worked opportunistically to yield diverse characteristics of neuronal tissue, from subcellular components to properties of ensembles, which converge on the required cognitive properties of WM. Priming and implicit memory may play supporting roles with WM. These intermediate-term memory phenomena allow certain kinds of background information to be accumulated at higher volume than seems possible from the textbook, "modal model" of memory. By expediting attentional focus on subsets of information already in long-term memory, priming may help WM chunks to emerge in limited number as appropriately scaled "figures" from the primed "ground." The larger neuronal dynamic patterns that embody these cognitive phenomena must regulate their microscopic component systems, automatically selecting those having parameters of temporal persistence, rhythm, and connectivity patterns that are pertinent to the current task. Relevant neural phenomena may include "Hebbian" associativity and persistence of firing patterns in prefrontal or hippocampal neurons. A conceivable basis for scaling and normalizing WM representations, along arbitrarily long or short ranges of any cognitive dimension, involves harmonic multiplier relationships among brain electrical rhythms and/or among topographical spatial periodic representations.

Human working memory capacity is 7+/-2 in a radial maze with distracting interruption: possible implication for neural mechanisms of declarative and implicit long-term memory.

Human participants were instructed to walk out along each of the arms of a 15-m in diameter, 8-arm radial maze once and only once. In order to approximate the circumstances under which laboratory rats remember visited sites, our human participants were asked to select arms in an unsystematic order. They scored an average of 7.6 to 7.8 correct choices, even if midway during a trial there was a 5-min interruption filled with a verbal-spatial interfering task (a scavenger hunt) or a 15-min interruption filled with a visuospatial task (a maze-running computer simulation). This finding extends our earlier research with humans in 13- or 17-arm radial mazes under nondelay conditions, in which we also found working memory (WM) capacity for about 7 to 9 places, the same as that of laboratory rats. We discuss earlier findings in other laboratories, showing that rats can successfully bridge long radial maze task interruptions of 5 or 8 h, and we compare our results also to those from studies in which human participants were not discouraged from reducing memory load by responding systematically in radial mazes. Because the radial maze task takes minutes to complete even under nondelay conditions its routine consideration as a working memory task in the animal literature alters the assumptions often made about the duration of WM in the human literature. Accumulating empirical findings about place-memory in humans, nonhuman mammals, and birds suggest it might be productive to reevaluate this theoretical issue with respect to present knowledge about the roles of the hippocampus and other brain structures in declarative memory and in procedural or implicit memory, while considering the hypothesis that some forms of information may exploit long-term memory in parallel with working memory.

Spatial working memory score of humans in a large radial maze, similar to published score of rats, implies capacity close to the magical number 7 +/- 22.

To compare the working memory (WM) capacity of humans to rats, we tested humans with a 17-arm radial maze and, in a follow up experiment, with a 13-arm radial maze. Both mazes were 15.2 meters in diameter, painted on a grassy field. In one version of the 13-arm experiment, we required a concurrent nonsense vocalization to impede subjects' use of language to remember locations. Subjects were instructed to choose arms of the radial maze unsystematically--as rats generally appear to do--and to visit the end of each arm only once. In additional procedures, we tested working memory capacity in a verbal task that is more analogous to the radial maze than is the typical ordered recall test. Subjects were asked to try to recite a sequence of 17 numbers (i.e., 18 through 34) or letters (A through Q) in unsystematic order, with no repeats. In another experiment subjects recited 13 numbers (14-26) or letters (A-M). In all tests, subjects were allowed only as many responses as there were distinct items (17 or 13, respectively). Average correct-response (nonrepeat) scores were 14.4 for the 17-arm maze and 14.1 for both of the verbal 17-item tests; these scores are close to the reported score for rats in a 17-arm radial maze. Average scores were between 10.8 and 11.4 in all of the 13-item maze and recitation tasks.(ABSTRACT TRUNCATED AT 250 WORDS)

Behavioral effects of SI versus SII cortex ablations on tactile orientation-localization and postural reflexes of rats.

Ten rats were observed, before and after selective unilateral ablation of the SI and/or SII cortical area, for the ability to turn toward tactile stimuli to head, limbs, or trunk. Placing and hopping postural reflexes were also observed. SI damage caused greater contralateral deficits than SII damage in all these behavioral measures; combined ablation of SI plus SII caused the greatest contralateral deficits in all measures. Partial recovery of all behaviors was observed during six postoperative test sessions, spanning 2 months. In contrast to these rats, cats studied in earlier experiments in this laboratory had shown a double dissociation, with SI damage followed by contralateral deficits in posture and movement but not passive touch, and damage to SII plus subjacent cortex by contralateral deficits in passive touch but only very small deficits in posture and movement. In other species comparisons, normal rats oriented as vigorously as do cats to tactile cues and, like cats, showed tactile responsiveness in orientation-localization greater rostrally than caudally. In both species wrong-way orientations were occasionally observed during the early postoperative period following the largest unilateral lesions. Unoperated albino rats oriented much less readily than do unoperated cats toward appetitive auditory or visual cues signaling availability of food. An accompanying theoretical review paper further examines the nature of passive and active touch in terms of these and other comparative findings.

Behavioral specializations of SI and SII cortex: a comparative examination of the neural logic of touch in rats, cats, and other mammals.

In what ways do passive touch and active touch require particular properties of cortical physiology? Because it is not easy to make predictions "bottom up" from neurophysiology and anatomy to behavior, there have been surprises in studying behavioral deficits following selective ablations of SI versus SII. For example, in earlier research on cats, unilateral ablation of SI unexpectedly had no effect on passive touch but did cause contralateral losses in posture and movement; it was ablation of SII plus subjacent cortex that led to contralateral losses in passive touch. In contrast, in a recent experiment with rats unilateral ablation of either SI or SII caused contralateral losses both in touch and postural reflexes. Some of the literature on dogs, monkeys, and humans suggests similarities with cats. To better understand these comparative findings, the following theoretical factors are considered: the degree of diffuseness versus specificity in brain input-output relations demanded or permitted by a behavior, the spatial and temporal scales of a behavior, the species' degree of encephalization, the need for stimulus generalization or functional equivalence of movements, and the relative sizes and sensitivities of different parts of the body. Hypotheses are also offered about why the evolution in rats of shortened forelimbs and increased vibrissal function may have entailed a peculiar compromise between active and passive touch and between functions of SI and SII.

Human performance with a seventeen-arm radial maze analog.

Results using radial mazes suggest rats may have a greater short-term memory (STM) capacity than the human "magical number seven." We examined spatial STM in humans using a radial maze analog drawn on paper. Subjects were instructed to lift, in "random" order, each of 17 cardboard flaps arranged radially around a center point. Fifteen undergraduate subjects, tested seven trials a day on 2 consecutive days, averaged 15.4 correct choices per trial. Thus, humans perform equally well on a 17-arm radial maze analog as rats do on a 17-arm radial maze, suggesting comparable spatial STM capacities, and perhaps homologous brain substrates for these tasks, in these two species.

An hypothesis about redundancy and reliability in the brains of higher species: analogies with genes, internal organs, and engineering systems.

The phenomena of behavioral resistance to massive brain damage and behavioral recovery from brain damage suggest there is redundancy in neural tissue. This paper uses basic concepts from probability theory and reliability engineering, as a first step toward more rigorously establishing the plausibility of the redundancy hypothesis. Exponential effects in the relevant formulas lead to results that are intuitively surprising. Thus, within a broad range of parametric assumptions related to lifespan and number of neurons or neural subsystems, it appears that the human brain may be at least twice as large as it would have to be for short-term survival. Simple reliability models suggest that redundancies are in parallel connections of smallest subsystems, such as individual neurons. Other implications of the basic formulas concern the relation between backed-up subsystem reliability and lifetime usage frequency for each subsystem, and the evolution of approximately equal allocation of lifetime reliability among components of a system. In addition, the paper briefly reviews more complex reliability engineering approaches. Redundancy as a reason for neural mass action is compared to other theoretical reasons for mass action in sensorimotor function and learning. Relationships of the present hypothesis to other theories of recovery from brain damage and to theories of regressive trophic phenomena in ontogeny are briefly discussed; it is suggested that as stages of ontogeny progress, both redundancy and flexibility in simpler behavioral functions are traded away for a larger, more differentiated repertoire of complex functions and memories.

Failure to find electrophysiological correlates of chronic neuroleptic-induced oral dyskinesias in cats: somatosensory and substantia nigra evoked potentials, electroencephalogram, and caudate spindles.

Each of nine cats was prepared with 30 chronically implanted stainless steel gross electrodes in cortex, basal ganglia and other brain structures. Measurements were taken for 0.5-11.5 months of baseline, chronic daily administration of chlorpromazine, and withdrawal. In most cases there was also a second cycle of drug administration and withdrawal. Although we observed dramatic, persistent increases in licking behavior, suggestive of tardive dyskinesia, consistent correlated patterns were not observed in somatosensory or substantia nigra evoked potentials, electroencephalogram, or spindling evoked in cortex by caudate stimulation.

Oral dyskinesia in brain-damaged rats withdrawn from a neuroleptic: implication for models of tardive dyskinesia.

Rats with ablated frontal sensorimotor cortex and one with ablated sensorimotor connections to forebrain showed more vacuous chewing movements following 6-week chronic administration of a neuroleptic than did occipitally damaged rats or normal controls who were treated in the same way. The effect was still present 1 month after withdrawal. It was not clearly enhanced by subsequent treatments. Other behaviors (e.g., walking, rearing, or grooming) were not similarly affected by drug withdrawal. Additional results of terminal probes with amphetamine, apormorphine, and haloperidol are described, including movements labeled 'sham eating', observed only in frontal rats given apomorphine (AP). The results are interpreted in terms of a Jacksonina model of levels of brain organization; such a model may be applicable to tardive dyskinesia, seen in many schizophrenic patients who are maintained on neuroleptics for long periods.

Does the brain actively maintain itself?

All living systems have special mechanisms for combatting entropy; however, the brain has dimensions of organized complexity beyong those manifest in the anatomical structure and physiology of the rest of the body. Reasons are given in support of the notion that the brain therefore must have a special, intrinsic "homeostatic" system for its information bearing structures, and, further, that slow electroencephalographic activity has properties which might make it useful for such an order-maintaining function. Recovery from brain damage is hypothesized to be a byproduct of this process, which may involve a cruder sort of information processing than occurs with such functions as perception and learning. Synchronized EEG activity may be adequate to handle this sort of information processing. Speculations are offered about possible mechanics, on the neuronal level, of slow wave participation in plasticity; for example, one such suggestion is based on findings that electrical fields can influence cellular orientation. The methodology of discovering the distribution within the brain of the hypothetical maintenance system is discussed briefly.

A neural systems theory of schizophrenia and tardive dyskinesia.

Some systems ideas applied to individual persons are used to try to explain symptoms of schizophrenia and a syndrome of uncontrolled fragments of movement which sometimes occurs as a side effect of chronic, antipsychotic drug therapy. The behavior of normal organisms may be conceptualized in three echelons of control, with each successively higher echelon organizing, by selective disinhibition, semiautonomous, spontaneous fragments of activity which comprise the next lower echelon. It is hypothesized that schizophrenia involves a deficiency of inhibition by the frontal cortex, first echelon, on the corpus striatum, second echelon. This results first in insufficiently integrated fragments of behavior, and second in premature associative linkages among active elements. First echelon control develops as a normal person matures and gradually loses some of the playful activities of childhood. It is hypothesized that by disrupting certain aspects of activity in the corpus striatum, neuroleptic drugs reduce schizophrenic symptoms but also reduce the capacity of the second echelon to inhibit and integrate the smaller behavioral fragments wired into lower parts of the brain, third echelon. This results in uncontrolled movements. Though many researchers already favor the hypothesis that neuroleptic drugs act on the corpus striatum, the broader theory presented here is new and depends in large part on general living systems considerations. Emphasis is on conceptual decomposition of the integrated behavior of a whole organism into less complex subsystems. Individually, these have neither too much nor too little complexity to yield a plausible model. Some experimental predictions and predictions about possible therapies are made from the theory.