figshare
Browse
Figure_2.tif (410.04 kB)

Prediction Accuracy and Average Response for Different Pattern Lengths in Both Tasks (Pattern, Random).

Download (0 kB)
figure
posted on 2013-02-20, 05:38 authored by Carolina Feher da Silva, Marcus Vinícius Chrysóstomo Baldo

The prediction accuracy and average response for different pattern lengths in both tasks: Pattern, when the sequence was formed by a repeating pattern, and Random, when the sequence was shuffled randomly. In different simulations, the animats had 4 (figure panel A) or 8 (figure panel B) hidden nodes. The error bars are the standard errors for n = 12. PM  =  Expected average response for animats that do probability matching. Max  =  Expected average response for animats that perseverate. Random PM  =  Expected accuracy for animats that do probability matching in the random task. Random Max  =  Expected accuracy for animats that perseverate in the random task.

History

Usage metrics

    PLOS ONE

    Keywords

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC