The applet requires Java 1.4.1 or higher. It will not run on Windows 95 or Mac OS 8 or 9. Mac users must have OS X 10.2.6 or higher and use a browser that supports Java 1.4. (Safari works, IE does not.)
created with NetLogo
view/download model file: 1DVariantTrading.nlogo
This simulation illustrates the ability of category competition and associated variant trading to preserve contrast between two categories.
The model follows the evolution of two one-dimensional categories in a production/perception loop where outputs are produced from each category and then restored in one of the two categories. The contents of each category is represented as a set of exemplars of previously categorized percepts (red and blue squares, respectively). In each round, several exemplars are randomly chosen from each of the categories in turn and an averaged output is produced. Gaussian noise added to outputs in production ensures that novel variants are constantly added to the system. The variant output is in turn compared to the average values of the two categories and stored as a new exemplar in the category with the closest mean. This is the crux of the simulation: if an output happens to be more similar to the other category, it will be stored as a new exemplar of that category, rather than of its own originating category, in a process I've termed 'variant trading'.
Slow memory decay is modeled through eventual deletion of older exemplars. Continual reproduction of exemplars in the form of new outputs and deletion of old exemplars allows the categories to evolve under the influence of any biases that may exist in production and perception.
The bias of interest here arises from variant trading between the two categories at their boundary, which encourages the categories to remain separate/contrastive as they evolve. If category competition is turned off by switching the button on the control panel of the simulation, each output will be stored back into its originating category and there is no pressure for the categories to occupy distinct regions of the parameter space.
Categories are represented in the model by red and blue square exemplars. Multiple exemplars that have the same value are stacked vertically, such that the image of the two categories on the screen is equivalent to the distribution of exemplar values in each category.
Click on set-up, and set 'Variant-trading' to 'off', which makes each output be stored back into the category it came from. You'll see a stack of red and blue squares at the center of the parameter dimension; at this starting point the two categories have identical contents. Click on 'go once' a few times and notice how the originally narrow distribution shared by both categories relaxes under the influence of noise in production. Also notice that the distribution of red and blue exemplars remains mixed. Now click set-up again to reset the simulation, and set 'Variant-trading' to 'on', which makes each output be stored in the closest category. Click on 'go once' a few times and notice how the two colors quickly segregate to separate sides of the overall distribution. The percent-overlap between the two categories over time is recorded in the graph at lower left. Now click on 'go' and watch the evolution of the two categories over time. Notice that the two categories always remain separate. To stop the simulation, click on 'go' again.
If you restart the simulation with 'Variant-trading' off and let it go for many cycles, you'll notice that while the two categories may be separate some of the time, they also spend some of the time occupying the same region of the parameter space. It is this approach and merger that is inhibited by variant trading under competition between the two categories.