COMPARATIVE COGNITION

Comparative cognition is the comparison of how animals (including humans) process and interact with the world. What sets animal cognition apart from the behavioristic tradition of stimulus-response (S-R) learning is the recognition that (mental) processes intervene between the stimulus and response. Some animal-cognitive procedures may be complex (e.g., "feeling of knowing" experiments by Hampton, 2001). But even "simple" Pavlovian conditioning procedures can deal with cognitive processes: for example, taste-aversion conditioning to the "image" of an expected type of food (e.g., Holland, 1990), or higher-order and backward fear conditioning to a neutral stimulus never paired with shock (Cole, Barnet, and Miller, 1995). Techniques developed in the late twentieth century to study cognitive processes have produced exciting revelations about how animals perceive, learn, remember, navigate, communicate, time, and count. This entry highlights some of these developments.

Interval Timing

A tremendous amount of information has accumulated since the 1980s about the interval-timing mechanisms of rats, pigeons, and primates. In a popular "peak" procedure, a stimulus cue is presented and after a fixed elapsed time the next response is rewarded. On some trials, reward is omitted and responding increases to a peak rate and then decreases. The peak indicates maximum expectancy of reward. Interval timing typically displays the scalar property—variability (spread) increases proportionally with time. The interval-timing theory receiving the most attention is shown in Figure 1. It includes a clock (pacemaker), a switch (gate) to an accumulator (integrator), reference memory for the expected time, and a comparator for the reference time versus accumulated time to see if "time's up." Converging evidence has been accumulating for each of these hypothetical processes. One example is that drug manipulations of the clock produce abrupt changes in peak time (see Figure 2) with gradual compensation by storage of new accumulator values in reference memory, whereas drug manipulations of the memory produce gradual changes (see Figure 3) due to distorted intervals accumulating in reference memory (Meck, 1996).

Numerosity: A Scale of Numbers

Learning a number scale is fundamental to mathematics. Evidence indicates that rhesus monkeys can learn a number scale (Brannon and Terrace, 1998, 2000). Monkeys touched randomly placed areas on a computer screen in order: first the area containing one object, then two, then three, and finally four objects. Objects varied in size, shape, or color and combinations of these dimensions within and across trials. The most remarkable result was that this training spontaneously generalized to larger novel numbers of five to nine objects beyond the range of training numbers.

Landmark Navigation and Cognitive Maps

Clark's nutcrackers apparently use directional bearings from several landmarks to "fix" the location of seed caches (Kamil and Jones, 2000; Kamil and Cheng, 2001). Nutcrackers were more accurate using multiple bearings than any single bearing with a distance measure along the bearing line. This remarkable finding came from manipulating the goal site in relationship to geometrical properties of fixed landmarks rather than manipulating the landmarks themselves. Landmarks seem to function as a configural whole, not as independent elements. Young children, but not older children or adults, use similar geometric navigation strategies (Hermer-Vazquez, Moffet, Munkholm, 2001). The issue of cognitive maps, however, is complex and is more than configural-landmark navigation (Shettleworth, 1998).

Abstract Concepts

Researchers have long debated which species are capable of abstract concept learning. Abstract concepts are rules (same versus different, identity matching) that transcend the training stimuli. Novel stimuli are used to test for abstract concept learning. Pigeons (like humans, apes, dolphins, and monkeys) are capable of learning abstract concepts. One study showed matching-to-sample (MTS) concept learning by pigeons (Wright, 1997). In MTS, a sample stimulus is followed by two choice stimuli, one of which is correct and matches the sample. Pigeons required to respond to (peck) the sample twenty times before getting the choice stimuli learned the concept fully (as opposed to partially). Other pigeons required to respond fewer times (e.g., once) did not learn the concept at all, and instead learned the configural pattern of each display (sample plus choice stimuli).

Pigeons are also capable of learning a same/different (S/D) abstract concept. In S/D, subjects identify stimuli as same or different. Pigeons most rapidly learn S/D (and the S/D concept partially) with large 384 element arrays and contrasting target regions on different trials (Cook, Katz, and Cavoto, 1997), or with sixteen-element arrays that are composed of all different or all same elements (Young, Wasserman, and Garner, 1997). But when the number of elements in these tasks is reduced to two (minimum number), performance falls to chance. Nevertheless, pigeons are capable of learning a S/D task when trained with two stimuli from the beginning and attain full S/D concept learning when the training set is expanded to 128 to 512 pictures. Rhesus monkeys attained full S/D concept learning with somewhat smaller set sizes than pigeons, and chimpanzees remarkably learned an S/D concept with only two training stimuli (Oden, Thompson, and Premack, 1988).

Equally remarkable is the S/D concept learning by a parrot, Alex, with stimuli that varied along three dimensions (Pepperberg, 1987). Alex responded with the English words "color," "shape," or "mah-mah" (matter). When Alex was presented with a red wooden triangle and a green rawhide triangle, for example, and asked, "What's same?" he typically said, "Shape." When presented with a red wooden square and a blue wooden square and asked, "What's different?" he typically said, "Color." Alex's novel object transfer performance was 85 percent correct. What makes this experiment so remarkable was that Alex had no way of telling which dimension (color, shape, or matter) would be questioned and whether he would be asked, "What's same?" or "What's different?"

Abstract Concepts Based on Categories

Research developments have shown that monkeys could learn an S/D concept with food versus non-food categories (Bovet and Vauclair, 2001). Baboons were given experience with the items and then presented with pairs of them (behind a window). They were required to pull one of two ropes to indicate whether they were of the same or different categories (food versus nonfood). Novel item transfer was better than 80 percent correct. Rhesus monkeys have also showed category matching (Neiworth and Wright, 1994).

Relations Between Relations: Analogical Reasoning

If abstract concept learning is higher order, then analogical reasoning is higher-higher order. The Miller's Analogies Test is a test of relations between relations. A recent MTS study showed that baboons could learn a relation between relations (Fagot, Wasserman, and Young, 2001). Baboons saw a brief sixteen-icon sample, which was then replaced by two sixteen-icon choice displays. Neither choice display had any icons in common with the sample. The baboons successfully learned and transferred this performance to novel stimuli, which could not be based on stimulus identity or functional categories. It is difficult to escape the conclusion that this behavior is based on a relation between relations. A similar conclusion came from a study on rhesus monkey music perception (Wright et al., 2000). Six-note tonal (childhood songs, tonal algorithm) musical passages, separated by one or two octaves were judged "same" better than 80 percent of the time. Tonal passages form a relation (tune), and thus this octave generalization (no common frequencies) is a relation between relations.

Great apes are particularly remarkable in their ability to form relations between relations. Chimpanzees with no relations-between-relations training "spontaneously" performed correct analogical choices without reward (Thompson, Oden, and Boysen, 1997). If anything could be more remarkable, Sarah, a highly trained chimpanzee arranged four out of five "scattered" geometric forms in the unique analogical relationship (and set aside the distractor) without explicit training (Thompson and Oden, 2000).

Episodic Memory

Episodic memory is memory for a particular event and is related to the issue of consciousness in animals. Recent experiments show that scrub jays have episodic memory (Clayton and Dickinson, 1998; Clayton and Dickinson, 1999; Emery and Clayton, 2001). In one experiment, jays cached wax worms and peanuts in separate and distinctive halves of two distinctive sand-filled ice cube trays (two-by-seven arrays). After four hours, the jays recovered the more desirable wax worms first from one tray. After 124 hours, they recovered the peanuts first from the other tray. In pretraining, they had learned that the wax worms deteriorated after 124 hours. Reversal of their earlier preference for wax worms means that they remembered what (peanuts versus wax worms), when (four versus 124 hours), and where (which tray side) the foods were stored. Cache recovery by Clark's nutcrackers in the forest must also be episodic memory because they use different cache sites each year and overcome potential proactive interference (i.e., confusions) from previous years' sites. Thus, they too know what, when, and where.

Feeling of Knowing: Metacognition

Exciting research developments with monkeys indicate that they "know when they remember" (Hampton, 2001). This ability in humans indicates conscious cognition and is closely aligned with the issue of consciousness and recognition of "self." In a MTS experiment, rhesus monkeys saw a sample (clip-art image), touched the sample, had a retention delay, and then decided whether or not to take a memory test. If they chose to take the memory test, four choices were presented on the four corners of the monitor and reward for the correct choice was highly desired peanuts. If they declined to take the test, they (automatically) received a less preferred pellet. The monkeys were more accurate when they chose to take the test than when they were required to take the test, and this difference increased with delay. An important aspect of this study was requiring the monkeys to monitor their memory before the test was presented. It is difficult to escape the conclusion that these monkeys know when they remember. Pigeons do not show this choice-test advantage (Inman and Shettleworth, 1999), and thus may lack metacognition.

Directions for Future Research

Animals have cognitive abilities that were previously thought to be uniquely human. Other qualitative differences may disappear as procedures better fit the predispositions of species being tested (Clayton and Dickinson, 1999). But there are issues other than finding human cognitive abilities in animals and these issues continue to revolve around what is meant by "comparative" and "cognition."

The Comparative Issue

Criticisms of comparative cognition have been that they are not comparative (Shettleworth, 1993). However, researchers have seen species comparisons in interval timing, landmark navigation, abstract concept learning, analogical reasoning, and metacognition. A problem with proving qualitative differences is that one can never be sure that some other test conditions would not show this ability. The study of quantitative differences encounters the same difficulty. Some argue that comparisons should be among closely related species to study cognitive evolution (Cambell and Hodos, 1991) or environmentally specialized cognitive behavior (Sherry and Schacter, 1987). But even testing closely related species (e.g., Clark's nut-crackers, Mexican jays, pinyon jays, and scrub jays) in the same task does not ensure that quantitative differences represent cognitive differences, as opposed to the task fitting the predispositions of some of the species better. Nevertheless, testing the species in a variety of tasks and varying critical parameters can converge on persuasive evidence for quantitative differences (Olson, Kamil, Balda, and Nims, 1995).

A more overarching goal would seem to understand the mechanisms and processes involved; in short, how cognition works. Any thorough understanding will need evidence from at least three domains: conditions (i.e., natural context) best suited to express this cognitive ability; laboratory tests that manipulate critical parameters over a substantial range; and neurobiology and neural circuitry that mediates this behavior. One can begin in any domain, but will need regular input from the other domains—a sort of cumulative upward spiral and a revisit of issues as evidence accumulates. Ultimately, quantitative differences (e.g., interval timing [Church, Meck, and Gibbon, 1994]) and even qualitative differences (e.g., serial-order learning [Terrace, Chen, and Newman, 1995]; memory rehearsal [Cook, Wright, and Sands, 1991]) should be shown to be individual points of comparison along much larger functional relationships revealed by parametric studies and quantitative models.

The Cognitive Issue

Issues associated with the cognition part of comparative cognition mainly revolve around the strength of evidence for private events (cognitive processes) between the stimulus and response. There will certainly be skeptics of the feeling-of-knowing experiments with animals (Hampton, 2001), but this is only the tip of the iceberg. It gets only better (or worse depending on one's persuasion) with experiments on beliefs, intentions, wants, desires, altruism, concept of self, and knowledge about another's beliefs and intentions—the so-called "theory of mind" experiments (Povinelli, 2001; Tomasello and Call, 1997). These investigations have stimulated a great deal of enthusiasm among scientists. Continued growth and acceptance will likely depend on parametric studies (i.e., manipulating critical variables over a substantial range) and converging evidence from several experiments like the blue jay and wax-worm studies. Furthermore, researchers will need to show that their findings cannot be accounted for by simpler associative learning principles involving observable stimuli and responses, like behavior-based theories timing (Killeen and Fetterman, 1988; Machado, 1997) as opposed to scalar-timing theory (see Figure 1). A wide variety of learning, cognitive, and memory phenomena can be accounted for by careful parametric experimentation coupled with powerful mathematical models to reveal underlying processes (Gallistel and Gibbon, 2000; White and Wixted, 1999). Any time we are tempted to get too heady over some new animalcognitive phenomenon, we might reflect on the exuberance originally generated by ape-language studies, and recall the instructive (if not tongue-in-cheek) S-R conditioning demonstrations of the pigeon's (cognitive) concepts of insight, symbolic communication, and self conducted by Epstein, Skinner, and their collaborators (1981).

Bibliography

Bovet, D., and Vauclair, J. (2001). Judgment of conceptual identity in monkeys. Psychonomic Bulletin & Review 8, 470-475.

Brannon, E. M., and Terrace, H. S. (1998). Ordering of the numerosities 1 to 9 by monkeys. Science 282, 746-749.

—— (2000). Representation of the numerosities 1-9 by rhesus macaques (Macaca mulatta). Journal of Experimental Psychology: Animal Behavior Processes 26, 31-49.

Campbell, C. B. G., and Hodos, W. (1991). The scala naturae revisited: Evolutionary scales and anagenesis in comparative psychology. Journal of Comparative Psychology 105, 211-221.

Church, R. M., Meck, W. H., and Gibbon, J. (1994). Application of scalar timing theory to individual trials. Journal of Experimental Psychology: Animal Behavior Processes 20, 135-155.

Clayton, N. S., and Dickinson, A. (1998). Episodic-like memory during cache recovery by scrub jays. Nature 395, 272-274.

—— (1999). Memory for the content of caches by scrub jays (Aphelocoma coerulescens). Journal of Experimental Psychology: Animal Behavior Processes 25, 82-91.

Cole, R. P., Barnet, R. C., and Miller, R. R. (1995). Temporal encoding in trace conditioning. Animal Learning and Behavior 23, 144-153.

Cook, R. G., Katz, J. S., and Cavoto, B. R. (1997). Pigeon same-different concept learning with multiple stimulus classes. Journal of Experimental Psychology: Animal Behavior Processes 23, 417-433.

Cook, R. G., Wright, A. A., and Sands, S. F. (1991). Interstimulus interval and viewing time effects in monkey list memory. Animal Learning and Behavior 19, 153-163.

Emery, N. J., and Clayton, N. S. (2001). It takes a thief to know a thief: Effects of social context on prospective caching strategies in scrub jays. Nature 414, 443-446.

Epstein, R., Lanza, R. P., and Skinner, B. F. (1981). "Self-awareness" in the pigeon. Science 212, 695-696.

Fagot, J., Wasserman, E. A., and Young, M. E. (2001). Discriminating the relation between relations: The role of entropy in abstract conceptualization by baboons (Papio papio) and humans (Homo sapiens). Journal of Experimental Psychology: Animal Behavior Processes 27, 316-328.

Gallistel, C. R., and Gibbon, J. (2000). Time, rate, and conditioning. Psychological Review 107, 289-344.

Hampton, R. R. (2001). Rhesus monkeys know when they remember. Proceeding of the National Academy of Sciences of the United States of America 98, 5,359-5,362.

Hermer-Vazquez, L. Moffet, A., and Munkholm, P. (2001). Language, space, and the development of cognitive flexibility in humans: The case of two spatial memory tasks. Cognition 79, 263-299.

Holland, P. C. (1990). Event representation in Pavlovian conditioning: Image and action. Cognition 37, 105-131.

Inman, A., and Shettleworth, S. J. (1999). Detecting metamemory in nonverbal subjects: A test with pigeons. Journal of Experimental Psychology: Animal Behavior Processes 25, 389-395.

Kamil, A. C., and Cheng, K. (2001). Way-finding and landmarks: The multiple-bearings hypothesis. Journal of Experimental Biology 204, 103-113.

Kamil, A. C., and Jones, J. E. (2000). Geometric rule learning by Clark's nutcrackers (Nucifraga columbiana). Journal of Experimental Psychology: Animal Behavior Processes 26, 439-453.

Katz, J. S., Wright, A. A., and Bachevalier, J. (2002). Same/different concept learning in Rhesus monkeys. Journal of Experimental Psychology: Animal Behavior Processes 28.

Killeen, P. R., and Fetterman, J. G. A behavioral theory of timing. Psychological Review 95, 274-295.

Machado, A. (1997). Learning the temporal dynamics of behavior. Psychological Review 104, 241-265.

Meck, W. H. (1996). Neuropharmacology of timing and time perception. Cognitive Brain Research 3, 227-242.

Neiworth, J. J., and Wright, A. A. (1994). Monkeys (Macaca mulatta) learn category matching in a nonidentical same-different task. Journal of Experimental Psychology: Animal Behavior Processes 20, 429-435.

Oden, D. L., Thompson, R. K., and Premack, D. (1988). Spontaneous transfer of matching by infant chimpanzees (Pan troglodytes). Journal of Experimental Psychology: Animal Behavior Processes 14, 140-145.

Olson, D. J., Kamil, A. C., Balda, R. P., and Nims, P. J. (1995). Performance of four seed-caching corvid species in operant tests of nonspatial and spatial memory. Journal of Comparative Psychology 109, 173-181.

Pepperberg, I. M. (1987). Acquisition of the same/different concept by an African Grey parrot (Psittacus erithacus): Learning with respect to categories of color, shape, and material. Animal Learning and Behavior 15, 423-432.

Povinelli, D. J. (2001). On the possibilities of detecting intentions prior to understanding them. In B. F. Malle, L. J. Moses, and D. A. Baldwin, eds., Intentions and intentionality: Foundations of social cognition. Cambridge, MA: MIT Press.

Sherry, D. F., and Schacter, D. L. (1987). The evolution of multiple memory systems. Psychological Review 94, 439-454.

Shettleworth, S. J. (1993). Where is the comparison in comparative cognition? Alternative research programs. Psychological Science 4, 179-184.

—— (1998). Cognition, evolution, and behavior. New York: Oxford University.

Terrace, H. S., Chen, S., and Newman, A. B. (1995). Serial learning with a wild card by pigeons (Columba livia): Effect of list length. Journal of Comparative Psychology 109, 1,162-1,172.

Thompson, R. K. R., and Oden, D. L. (2000). Categorical perception and conceptual judgments by nonhuman primates: The paleological monkey and the analogical ape. Cognitive Science 24, 363-396.

Thompson, R. K. R., Oden, D. L., and Boysen, S. T. (1997). Language-naïve chimpanzees (Pan troglodytes) judge relations between relations in a conceptual matching-to-sample task. Journal of Experimental Psychology: Animal Behavior Processes 23, 31-43.

Tomasello, M., and Call, J. (1997). Primate cognition. New York: Oxford University Press.

White, K. G., and Wixted, J. T. (1999). Psychophysics of remembering. Journal of the Experimental Analysis of Behavior 71, 91-113.

Wright, A. A. (1997). Concept learning and learning strategies. Psychological Science 8, 119-123.

Wright, A. A., Rivera, J. J., Hulse, S. H., Shyan, M., and Neiworth, J. J[GU2]. (2000). Music perception and octave generalization in rhesus monkeys. Journal of Experimental Psychology: General 129, 291-307.

Young, M. E., Wasserman, E. A., and Garner, K. L. (1997). Effects of number of items on the pigeon's discrimination of same from different visual displays. Journal of Experimental Psychology: Animal Behavior Processes 23, 491-501.

Anthony A.Wright