Berg's Card Sorting Task
A lot of research on human problem solving focuses on how we develop strategies to solve fixed problems. But problems often are not static; the problem changes, old strategies no longer work, and new strategies must be adopted. How do we adapt to changing problems? Grant and Berg (1948) developed a sorting task (now often called the Wisconsin Card Sort Test or WCST; Grant & Berg, 1948), used to investigate different response strategies and how they are developed (Eling, Derckx & Maes, 2008).
In the original version of the WCST, participants are seated at a table across from the experimenter. They are instructed to pick the top card from a 128-card deck, and to match the card to one of four face-up cards on the table. Each card depicts some number of shapes, and the drawn card could match the exemplars based on any of three different attributes: color (e.g, a card with three red squares matches one with two red circles), number (e.g., a card with three red squares matches one with three yellow diamonds), or shape (a card with three red squares matches one with two green squares). Because participants are not told which attribute to sort on, they must begin by guessing randomly and updating their beliefs about how to sort based on feedback from the experimenter.
The "catch" of the card sorting task is that the sorting rule changes during the test. In the original version, the sort rule was changed by the experimenter after ten cards had been sorted; in newer versions the rule changes after five correct card sorts in a row. When a participant goes from knowing that color is the correct sorting attribute to suddenly being told they are wrong, they have to update their internal rule and strategy (Nyhus & Barcelo, 2009). Over the course of the experiment, researchers note how many cards are successfully sorted, how many errors are made, and how often participants perseverate with an old sorting rule after the rule changes (patients with frontal lobe dysfunction may have difficulty disengaging from an irrelevant rule, resulting in what are called "perseverative errors"; Greve et al., 2002).
The WCST was first used in a clinical setting to assess disorders in the prefrontal lobe (Milner, 1963). Though the test is still used today to assess neurologic dysfunctions, most neuropsychological assessment guidelines warn against using WCST scores as an indicator of region-specific brain damage, especially to the frontal lobe (Drewe, 1974).
Short Version (default) This version of the card sort task uses a modified 24-card deck in which none of the cards match the sort piles on more than one attribute (i.e., number, shape, or color). Rule changes are made after five correct sorts in a row, and participants complete 64 total sorts. Original Version This version of the card sort task uses the full 64-card deck that contains all possible combinations of number, shape, and color. Rule changes are made after 10 correct sorts in a row, and participants complete 128 total sorts.
Drewe, E.A. (1974). The effect of type and area of brain lesion on Wisconsin card sorting test performance. Cortex, 10, 159-170.
Eling, P., Derckx, K., & Maes, R. (2008). On the historical and conceptual background of the Wisconsin Card Sorting Test. Brain and Cognition, 67, 247-253.
Grant, D.A., & Berg, E. (1948). A behavioral analysis of degree of reinforcement and ease of shifting to new responses in Weigl-type card-sorting problem. Journal of Experimental Psychology, 38, 404-411.
Greve, K.W., Love, J.M., Sherwin, E., Mathias, C.W., Houston, R.J., & Brennan, A. (2002). Temporal stability of the Wisconsin Card Sorting Test in a chronic traumatic brain injury sample. Assessment, 9, 271-277.
Milner, B. (1963). Effects of different brain lesions on card sorting: The role of the frontal lobes. Archives of Neurology, 9, 100-110.
Nyhus, E., & Barcelo, F. (2009). The Wisconsin card sorting test and the cognitive assessment of prefrontal executive functions: a critical update. Brain and Cognition, 71, 437-451.