Academic journal article Attention, Perception and Psychophysics

Criterial Noise Effects on Rule-Based Category Learning: The Impact of Delayed Feedback

Academic journal article Attention, Perception and Psychophysics

Criterial Noise Effects on Rule-Based Category Learning: The Impact of Delayed Feedback

Article excerpt

Variability in the representation of the decision criterion is assumed in many category-learning models, yet few studies have directly examined its impact. On each trial, criterial noise should result in drift in the criterion and will negatively impact categorization accuracy, particularly in rule-based categorization tasks, where learning depends on the maintenance and manipulation of decision criteria. In three experiments, we tested this hypothesis and examined the impact of working memory on slowing the drift rate. In Experiment 1, we examined the effect of drift by inserting a 5-sec delay between the categorization response and the delivery of corrective feedback, and working memory demand was manipulated by varying the number of decision criteria to be learned. Delayed feedback adversely affected performance, but only when working memory demand was high. In Experiment 2, we built on a classic finding in the absolute identification literature and demonstrated that distributing the criteria across multiple dimensions decreases the impact of drift during the delay. In Experiment 3, we confirmed that the effect of drift during the delay is moderated by working memory. These results provide important insights into the interplay between criterial noise and working memory, as well as providing important constraints for models of rule-based category learning.

Category learning is the process of establishing the memory traces necessary to organize objects and events in the environment into separate classes. Researchers have long debated the existence and qualitative nature of various category-learning systems. If any consensus has emerged from this debate, it is that categorization can occur by a process of explicit hypothesis testing (Allen & Brooks, 1991; Ashby, Alfonso-Reese, Turken, & Waldron, 1998; Erickson & Kruschke, 1998; Folstein & Van Petten, 2004; Nosofsky, Palmeri, & McKinley, 1994; Regehr & Brooks, 1995). According to the multiple learning systems perspective, a hypothesis-testing system would not be ideally suited for all category-learning tasks. Instead, a hypothesis-testing system is thought to be primarily responsible for learning rule-based tasks where accuracy is maximized by first learning which stimulus dimensions are relevant and then learning the placement of decision criteria along the relevant dimensions (Ashby et al., 1998; Ashby & Ell, 2001). Not surprising, computational models implementing a hypothesis-testing system have focused on how the decision criteria are updated in response to trial-by-trial information (e.g., the stimuli, corrective feedback; Ashby et al., 1998; Busemeyer & Myung, 1992; Erickson & Kruschke, 1998; Kac, 1962; Kubovy & Healy, 1977; Maddox, 2002; E. A. C. Thomas, 1973). Many models of categorization and decision making also emphasize the effect of variability in the representation of the decision criterion, or criterial noise (e.g., Ashby et al., 1998; Ashby & Lee, 1993; Benjamin, Diaz, & Wee, 2009; Dorfman, Saslow, & Simpson, 1975; Erev, 1998; Mueller & Weidemann, 2008; Treisman & Williams, 1984). Although criterial noise is thought to have a negative effect on categorization accuracy, few studies have examined how criterial noise effects might interact with the demands of rule-based category-learning tasks.

To begin, consider the categories in Figure 1A. Each point represents the spatial frequency and spatial orientation of a Gabor pattern (i.e., a sine wave grating in which contrast is modulated by a circular Gaussian filter). The optimal strategy involves learning the single unidimensional (1UD) decision criterion denoted by the solid vertical line. On each trial, the participant is presented with a single Gabor pattern and is instructed to assign the stimulus to Category A or B. Corrective feedback is provided immediately following the response, and the participant uses this feedback to learn the correct category assignments through trial and error. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.