Trait-like Features of Rumination

The next question that was addressed for human data is whether depression or the correlation between affective and semantic performance better explains variation in the network's performance on the lexical decision task. The answer to this question provides an interesting comparison between strictly statistical and causal modeling perspectives. Statistically, the correlation between valence identification and lexical decision reaction times was, as shown in Table 12, p. 116, very low. Depression did not explain a statistically significant amount of variation in any of the lexical decision variables, due to the very small observed effect. The relationship between the lexical decision variables and valence identification variables were sometimes greater than that between depression itself and performance on either task. Yet, the underlying explanation for this fact is not contained in the statistics. Performance on the valence identification task is dependent on performance on the lexical decision task and vice-versa are due to feedback within the affective semantic loop. Overtraining on some negative stimulus, or the induction of depression serves to change the feedback within the affective-semantic loop somewhat. Still, the primary effect of the initial training (potentially akin to premorbid functioning) is not erased by differential training on negativity. Given the small effect of negative overtraining on simulated lexical decision task performance, none of the hierarchical regressions using negative overtraining and valence identification variables as the independent variables and lexical decision variables as dependent variables were statistically significant, or accounted for more than 3$\%$ of the variation in the simulated lexical decision reaction times.

 

Table: Correlations Between Overtraining on Negative Exemplars (``Depression'') and Performance on the Simulated Lexical Decision and Valence Identification Tasks

	Depression	LD positive	LD negative	LD neutral
Depression	1	-.08	.15	.05
VID-positive	.57*	.05	.08	.04
VID-negative	-.24*	-.21*	-.18	.01
VID-neutral	-.01	.03	-.01	.02