nosof94exalcove: Simulation of CIRP nosof94 with ex-ALCOVE model

In catlearn: Formal Psychological Models of Categorization and Learning

Simulation of CIRP nosof94 with ex-ALCOVE model

Description

Runs a simulation of the nosof94 CIRP using the slpALCOVE model implementation as an exemplar model and nosof94train as the input representation.

Usage

  nosof94exalcove(params = NULL)

Arguments

params

A vector containing values for c, phi, la, and lw, in that order, e.g. params = c(2.1, 0.6, 0.09, 0.9). See slpALCOVE for an explanation of these parameters. Where params = NULL, best-fitting parameters are derived from optimzation archive nosof94exalcove_opt

Details

N.B.: This simulation uses a standard version of ALCOVE. For a replication of the ALCOVE simulation of these data reported by Nosofsky et al. (1994), which is non-standard in a number of respects, see nosof94bnalcove.

An exemplar-based simulation using slpALCOVE and nosof94train. The co-ordinates for the radial-basis units are assumed, and use the same binary representation as the abstract category structure.

Other parameters of slpALCOVE are set as follows: r = 1, q = 1, initial alpha = 1/3, initial w = 0. These values are conventions of modeling with ALCOVE, and should not be considered as free parameters. They are set within the nosof88exalcove function, and hence can't be changed without re-writing the function.

This simulation is reported in Wills & O'Connell (n.d.).

Value

A matrix of predicted response probabilities, in the same order and format as the observed data contained in nosof94.

Author(s)

Andy Wills

References

Nosofsky, R.M., Gluck, M.A., Plameri, T.J., McKinley, S.C. and Glauthier, P. (1994). Comparing models of rule-based classification learning: A replication and extension of Shepaard, Hovland, and Jenkins (1961). Memory and Cognition, 22, 352–369

Wills, A.J. & O'Connell (n.d.). Averaging abstractions. Manuscript in preparation.

See Also

nosof94, nosof94oat, nosof94train, slpALCOVE, nosof94bnalcove

catlearn documentation built on April 4, 2023, 5:12 p.m.