catlearn (version 0.8)

nosof88: Instantiation frequency CIRP

Description

Instantiation frequency is the number of times a stimulus has been observed as a member of a specific category (Barsalou, 1985). Increasing instantiation frequency of a stimulus increases categorization accuracy for that stimulus ('direct' effect), and for other similar stimuli ('indirect' effect). Experiment 1 of Nosofsky (1988) provides the data for this CIRP.

Usage

data(nosof88)

Arguments

Format

A data frame with the following columns:

cond

Experimental condition, see 'details'. 1 = 'B', 2 = 'E2', 3 = 'E7'

stim

Stimulus number, see Nosofsky (1988), Figure 1. Takes values: 1-12

c2acc

Mean probability, across participants, of responding that the item belongs to category 2.

Details

Wills et al. (n.d.) discuss the derivation of this CIRP. In brief, both the direct and indirect effects have been independently replicated. Experiment 1 of Nosofsky (1988) was selected due to the availability of a multidimensional scaling solution for the stimuli, see nosof88train.

Experiment 1 of Nosofsky(1988) involved the classification of Munsell chips of fixed hue (5R) varying in brightness (value) and saturation (chroma). Instantiation frequency was manipulated between subjects. In condition B, all stimuli were equally frequent. In condition E2 (E7), stimulus 2 (7) was approximately five times as frequent as each of the other stimuli. In condition E2 (E7), stimulus 4 (9) indexes the indirect effect. There were three blocks of training. Block length was 48 trials for condition B and 63 trials for conditions E2 and E7. The training phase was followed by a transfer phase, which is not included in this CIRP (see Nosofsky, 1988, for details).

The data are as shown in Table 1 of Nosofsky (1988). The data are mean response probabilities for each stimulus in the training phase, averaged across blocks and participants.

References

Barsalou, L.W. (1985). Ideals, central tendency, and frequency of instantiation as determinants of graded structure in categories. Journal of Experimental Psychology: Learning, Memory & Cognition, 11, 629-654.

Wills et al. (n.d.). Benchmarks for category learning. Manuscript in preparation.

See Also

nosof88train, nosof88oat