Alexandria C. Zakrzewski

Deferred Feedback Sharply Dissociates Implicit and Explicit Category Learning

The controversy over multiple category-learning systems is reminiscent of the controversy over multiple memory systems. Researchers continue to seek paradigms to sharply dissociate explicit category-learning processes (featuring category rules that can be verbalized) from implicit category-learning processes (featuring learned stimulus-response associations that lie outside declarative cognition). We contribute a new dissociative paradigm, adapting the technique of deferred-rearranged reinforcement from comparative psychology. Participants learned matched category tasks that had either a one-dimensional, rule-based solution or a multidimensional, information-integration solution. They received feedback either immediately or after each block of trials, with the feedback organized such that positive outcomes were grouped and negative outcomes were grouped (deferred-rearranged reinforcement). Deferred reinforcement qualitatively eliminated implicit, information-integration category learning. It left intact explicit, rule-based category learning. Moreover, implicit-category learners facing deferred-rearranged reinforcement turned by default and information-processing necessity to rule-based strategies that poorly suited their nominal category task. The results represent one of the strongest explicit-implicit dissociations yet seen in the categorization literature.

The time course of explicit and implicit categorization

Contemporary theory in cognitive neuroscience distinguishes, among the processes and utilities that serve categorization, explicit and implicit systems of category learning that learn, respectively, category rules by active hypothesis testing or adaptive behaviors by association and reinforcement. Little is known about the time course of categorization within these systems. Accordingly, the present experiments contrasted tasks that fostered explicit categorization (because they had a one-dimensional, rule-based solution) or implicit categorization (because they had a two-dimensional, information-integration solution). In Experiment 1, participants learned categories under unspeeded or speeded conditions. In Experiment 2, they applied previously trained category knowledge under unspeeded or speeded conditions. Speeded conditions selectively impaired implicit category learning and implicit mature categorization. These results illuminate the processing dynamics of explicit/implicit categorization.

The interplay between uncertainty monitoring and working memory: Can metacognition become automatic?

The uncertainty response has grounded the study of metacognition in nonhuman animals. Recent research has explored the processes supporting uncertainty monitoring in monkeys. It has revealed that uncertainty responding, in contrast to perceptual responding, depends on significant working memory resources. The aim of the present study was to expand this research by examining whether uncertainty monitoring is also working memory demanding in humans. To explore this issue, human participants were tested with or without a cognitive load on a psychophysical discrimination task that included either an uncertainty response (allowing the participant to decline difficult trials) or a middle-perceptual response (labeling the same intermediate trial levels). The results demonstrated that cognitive load reduced uncertainty responding, but increased middle responding. However, this dissociation between uncertainty and middle responding was only observed when participants either lacked training or had very little training with the uncertainty response. If more training was provided, the effect of load was small. These results suggest that uncertainty responding is resource demanding, but with sufficient training, human participants can respond to uncertainty either by using minimal working memory resources or by effectively sharing resources. These results are discussed in relation to the literature on animal and human metacognition.

Cross-modal information integration in category learning

An influential theoretical perspective describes an implicit category-learning system that associates regions of perceptual space with response outputs by integrating information preattentionally and predecisionally across multiple stimulus dimensions. In this study, we tested whether this kind of implicit, information-integration category learning is possible across stimulus dimensions lying in different sensory modalities. Humans learned categories composed of conjoint visual–auditory category exemplars comprising a visual component (rectangles varying in the density of contained lit pixels) and an auditory component (in Exp. 1, auditory sequences varying in duration; in Exp. 2, pure tones varying in pitch). The categories had either a one-dimensional, rule-based solution or a two-dimensional, information-integration solution. Humans could solve the information-integration category tasks by integrating information across two stimulus modalities. The results demonstrated an important cross-modal form of sensory integration in the service of category learning, and they advance the field’s knowledge about the sensory organization of systems for categorization.

Categorization: The View from Animal Cognition

Exemplar, prototype, and rule theory have organized much of the enormous literature on categorization. From this theoretical foundation have arisen the two primary debates in the literature—the prototype-exemplar debate and the single system-multiple systems debate. We review these theories and debates. Then, we examine the contribution that animal-cognition studies have made to them. Animals have been crucial behavioral ambassadors to the literature on categorization. They reveal the roots of human categorization, the basic assumptions of vertebrates entering category tasks, the surprising weakness of exemplar memory as a category-learning strategy. They show that a unitary exemplar theory of categorization is insufficient to explain human and animal categorization. They show that a multiple-systems theoretical account—encompassing exemplars, prototypes, and rules—will be required for a complete explanation. They show the value of a fitness perspective in understanding categorization, and the value of giving categorization an evolutionary depth and phylogenetic breadth. They raise important questions about the internal similarity structure of natural kinds and categories. They demonstrate strong continuities with humans in categorization, but discontinuities, too. Categorization’s great debates are resolving themselves, and to these resolutions animals have made crucial contributions.

Ecology, Fitness, Evolution New Perspectives on Categorization

Categorization’s great debate has weighed single-system exemplar theory against the possibility of alternative processing systems. We take an evolutionary and fitness perspective toward this debate to illuminate it in a new way. There are continuities in category processes—extending across millions of years in vertebrate evolution—that have profound theoretical implications. Thus, animals are crucial behavioral ambassadors to this area. They reveal the roots of human categorization, the basic assumptions of vertebrates entering category tasks, and the surprising weakness of exemplar memory as a category-learning strategy. These results have joined neuroscience results to prompt important changes in categorization theory. Categorization’s great debate is ending. Broad-based converging evidence now makes it clear that the unitary exemplar view is insufficient. Categorization is served by multiple systems of process and representation.

Decision deadlines and uncertainty monitoring: the effect of time constraints on uncertainty and perceptual responses.

The behavioral uncertainty response has grounded the study of animal metacognition and influenced the study of human psychophysics. However, the interpretation of this response is debated—especially whether it is a behavioral index of metacognition. The authors advanced this interpretation using the dissociative technique of response deadlines. Uncertainty responding, if it is higher level or metacognitive, should depend on a slower, more controlled decisional process and be more vulnerable to time constraints. Humans performed sparse–uncertain–dense or sparse–middle–dense discriminations in which, respectively, they could decline difficult trials or positively identify middle stimuli. Uncertainty responses were sharply and selectively reduced under a decision deadline, as compared to primary perceptual responses (i.e., “sparse,” “middle,” and “dense” responses). This dissociation suggests that the uncertainty response does reflect a higher-level, decisional response. It grants the uncertainty response a distinctive psychological role in its task and encourages an interpretation of this response as an elemental behavioral index of uncertainty that deserves continuing research.

I scan, therefore I decline: The time course of difficulty monitoring in humans (homo sapiens) and macaques (macaca mulatta).

The study of nonhumans’ metacognitive judgments about trial difficulty has grown into an important comparative literature. However, the potential for associative-learning confounds in this area has left room for behaviorist interpretations that are strongly asserted and hotly debated. This article considers how researchers may be able to observe animals’ strategic cognitive processes more clearly by creating temporally extended problems within which associative cues are not always immediately available. We asked humans and rhesus macaques to commit to completing spatially extended mazes or to decline completing them through a trial-decline response. The mazes could sometimes be completed successfully, but other times had a constriction that blocked completion. A deliberate, systematic scanning process could preevaluate a maze and determine the appropriate response. Latency analyses charted the time course of the evaluative process. Both humans and macaques appeared, from the pattern of their latencies, to scan the mazes through before committing to completing them. Thus monkeys, too, can base trial-decline responses on temporally extended evaluation processes, confirming that those responses have strategic cognitive-processing bases in addition to behavioral-reactive bases. The results also show the value of temporally and spatially extended problems to let researchers study the trajectory of animals’ online cognitive processes.

The Transfer of Category Knowledge by Macaques (Macaca mulatta) and Humans (Homo sapiens).

Cognitive psychologists distinguish implicit, procedural category learning (stimulus–response associations learned outside declarative cognition) from explicit-declarative category learning (conscious category rules). These systems are dissociated by category learning tasks with either a multidimensional, information-integration (II) solution or a unidimensional, rule-based (RB) solution. In the present experiments, humans and two monkeys learned II and RB category tasks fostering implicit and explicit learning, respectively. Then they received occasional transfer trials—never directly reinforced—drawn from untrained regions of the stimulus space. We hypothesized that implicit-procedural category learning—allied to associative learning—would transfer weakly because it is yoked to the training stimuli. This result was confirmed for humans and monkeys. We hypothesized that explicit category learning—allied to abstract category rules—would transfer robustly. This result was confirmed only for humans. That is, humans displayed explicit category knowledge that transferred flawlessly. Monkeys did not. This result illuminates the distinctive abstractness, stimulus independence, and representational portability of humans’ explicit category rules.