Brendan McGonigle

Are monkeys logical

THE monkeys status as a thinker has never been high; yet laboratory investigations testify, nevertheless, to the ability of many species of monkey to learn complex tasks, if not to reason. On this latter point, however, hard evidence is significantly lacking. One reason for this is that it is difficult to devise tests which are both meaningful to non-verbal subjects yet satisfy the stringent requirements of a formal reasoning test such as one adapted from Burt1 which first gives the subject the following information: “Edith is fairer than Suzanne”, “Edith is darker than Lili”, and then requires solution of the question, “which is the darkest, Edith, Suzanne or Lili?”. Bryant and Trabasso2 have devised a simplified method of giving such tests to very young children, and we have adapted this into a non-verbal one for use with monkeys.

Long-term retention of single and multistate prismatic adaptation by humans.

SENSORY rearrangement techniques have been used to study the way in which primates reach accurately for objects. When an object is viewed through wedge prisms, for example, reaching is disrupted but rapidly corrected when the subject is given immediate feedback concerning the position of his pointing limb relative to the target1. Such adaptation to distorted input is conventionally regarded as a process of minimising the discrepancy between two or more sensory channels (for example, vision and proprioception), with the neglected or non-dominant channel conforming to the position information provided by the other. The role of learning in this process is unclear. When the prisms are removed, the effects of such adaptation disappear rapidly (unlike most learning phenomena), and after an interval of 24 h during which the subject reaches ‘normally’, most investigators assume2 that all traces of adaptation have been eliminated from the nervous system. This view has been challenged recently by results we obtained with squirrel monkeys3, which retained their adaptation over periods often considerably longer than the 48 h spent in their home cages. We also found that they could adapt to two prisms displacing in equal and opposite directions, and eventually conserve both adaptations together with their normal reaching behaviour. Their progressive elimination of error (due to prismatic distortion) as a function of the number of occasions on which they were exposed to prism conditions suggests, furthermore, powerful motor learning mechanisms at work in the adaptation process. We now report similar findings with human subjects.

Agent Architecture as Object Oriented Design

Improving the development of agent intelligence requires improving the mechanisms of that development. This paper explores the application of an established software methodology, object-oriented design, to agent development in two ways. We present a distributed agent architecture, Edmund, and describe first its own object-oriented structure. Then we relate the methodology for developing agent behaviors under Edmund. We explain how this methodology exploits key aspects of object-oriented design, particularly the development of the class hierarchy, as a prototype for agent design.

Are Children Any More Logical Than Monkeys on the Five-Term Series Problem?.

Abstract Six-year-old children were tested on several versions of the five-term transitivity problem as used by B. O. McGonigle and M. Chalmers (1977, Nature (London) , 267 , 694–696) with squirrel monkeys as subjects. Both binary and triadic versions of the tests were administered in both verbal and nonverbal modes to help determine whether or not any major procedural differences between the monkey version and that used conventionally in research with children might account for the monkeys apparently nonlogical solution of the problem. The main result is that children showed very similar response profiles to that of monkeys in all the conditions used. In addition, “labeling”, direct seriation, and “association” post-tests suggest that nonlogical strategies can underwrite ostensibly impeccable transitive “reasoning” in child as well as monkey.

The learning of hand preferences by Squirrel Monkey.

SummaryIn a reaching task with Squirrel Monkeys stable hand preferences were established by reward training. The use of a particular hand was brought under the control of a colour cue.

Concurrent disjoint and reciprocal classification by Cebus apella in seriation tasks: evidence for hierarchical organization

We report the results of a 4-year-long study of capuchin monkeys (Cebus apella ) on concurrent three-way classification and linear size seriation tasks using explicit ordering procedures, requiring subjects to select icons displayed on touch screens rather than manipulate and sort actual objects into groups. The results indicate that C. apella is competent to classify nine items concurrently, first into three disjoint classes where class exemplars are identical to one another, then into three reciprocal classes which share common exemplar (size) features. In the final phase we compare the relative efficiency of executive control under conditions where both hierarchical and/or linear organization can be utilized. Whilst this shows a superiority of categorical based size seriation for a nine item test set suggesting an adaptive advantage for hierarchical over linear organization, Cebus nevertheless achieved high levels of principled linear size seriation with sequence lengths not normally achieved by children below the age of six years.

Serial Adaptation to Conflicting Prismatic Rearrangement Effects in Monkey and Man

Seven adult male squirrel monkeys adapted successfully to prisms displacing in equal and opposite directions. After protracted training under certain conditions, they succeeded in conserving adaptation to both prisms without benefit of readaptation trials. Results obtained with thirty-six undergraduates under similar conditions of test (although for a much shorter period) are also reported.

Levels of stimulus processing by the squirrel monkey: relative and absolute judgements compared.

Nine squirrel monkeys were required to select from various sets of stimuli—differing in size or brightness—either in terms of relational criteria or on an absolute stimulus basis. The level of information processing required by each task was assessed by means of stimulus transformation techniques, variations in set size, and by the elimination of the visible context. It was found that some relational judgements make fewer processing demands on the subject than do absolute stimulus judgements; the ‘middle’ relation, however, appears much more difficult to use than selection of a stimulus on an absolute basis and may be beyond the competence of the squirrel monkey. The results are seen as support for the thesis as advanced by McGonigle and Jones that the criteria of judgement, when varied, change the depth of stimulus processing by monkey as well as man.

The Perception of Linear Gestalten by Rat and Monkey: Sensory Sensitivity or the Perception of Structure?:

Four experiments on rats and squirrel monkeys are reported which show that the well-known transposition by animals to continuous from broken or interrupted line stimuli, first reported by Krechevsky, is attributable to their failure to transfer from simultaneous to successive discrimination of dot patterns. When given appropriate successive discrimination training, however, monkeys reverse their original preference and select dot instead of continuous line stimuli.

The Growth of Cognitive Structure in Monkeys and Men

There is a widespread view that the sorts of animal learning mechanisms most frequently studied in the laboratory are inductively too weak and unproductive to generate the kinds of behaviours expressed in higher order forms of human cognitive and linguistic adaptation (Chomsky, 1980; Fodor & Pylyshyn, 1988; Piaget, 1971). One reason for this (Harlow, 1949) is that investigations are rarely followed through from one learning episode to another to assess the cumulative benefits (if any) as a function of the agent’s task and life history. Yet the course of human development is protracted, and even sophisticated adult subjects frequently show dramatic changes in strategy when confronted with many problems of the same type, detecting pattern and structural invariance in some (e.g., Wood, 1978), using analogies to bridge problems of a different surface structure (Gentner, 1983), and devising progressively economic, data reducing procedures to secure success with the least investment in resource (Anderson, 1990; McGonigle & Chalmers, 1996, McGonigle & Chalmers, 1998; McGonigle & Chalmers, 2001). Whilst Harlow’s pioneering work on the learning set (LS) using primates is an exception, indicating the vast potential for accelerated learning, his generative claim that the LS leaves the organism free to attack problems of a new hierarchy of difficulty has never been properly realised due to the fact that all problems in conventional LS studies are of the same (simple binary) type and level of difficulty (but see Terrace this volume).