“Virus and Epidemic”: Causal Knowledge Activates Prediction Error Circuitry

Abstract

Knowledge about cause and effect relationships (e.g., virus-epidemic) is essential for predicting changes in the environment and for anticipating the consequences of events and one's own actions. Although there is evidence that predictions and learning from prediction errors are instrumental in acquiring causal knowledge, it is unclear whether prediction error circuitry remains involved in the mental representation and evaluation of causal knowledge already stored in semantic memory. In an fMRI study, participants assessed whether pairs of words were causally related (e.g., virus-epidemic) or noncausally associated (e.g., emerald-ring). In a second fMRI study, a task cue prompted the participants to evaluate either the causal or the noncausal associative relationship between pairs of words. Causally related pairs elicited higher activity in OFC, amygdala, striatum, and substantia nigra/ventral tegmental area than noncausally associated pairs. These regions were also more activated by the causal than by the associative task cue. This network overlaps with the mesolimbic and mesocortical dopaminergic network known to code prediction errors, suggesting that prediction error processing might participate in assessments of causality even under conditions when it is not explicitly required to make predictions.

Electrophysiological difference between the representations of causal judgment and associative judgment in semantic memory

“…The difference waves mainly activated the central and central-parietal sites. These results suggested that the representation of causal judgment involved more cognitive resources than associative judgment, complementary with behavior and fMRI studies (Fenker et al, 2010;Fugelsang & Dunbar, 2005;Satpute et al, 2005). That is, people appeared to distinguish the roles of cause and effect or bind the relevant events into the 'cause' and 'effect' roles in the representation of causal relations, but such distinction was not found in the representation of associative relations .…”

“…Following this, these researchers also investigated the neural basis of causal relations via fMRI (Fenker et al, 2010;Satpute et al, 2005). For example, Satpute et al (2005) found that causal judgments, in contrast to associative judgments, recruited greater activation in the left dorsolateral prefrontal cortex and the right precuneus related to working memory and reasoning.…”

“…For example, Satpute et al (2005) found that causal judgments, in contrast to associative judgments, recruited greater activation in the left dorsolateral prefrontal cortex and the right precuneus related to working memory and reasoning. Recently, Fenker et al (2010) requested participants to determine whether the word pair was consistent with task cue (causal or associative) shown at the beginning of each trial in one task, and asked them to assess whether the word pair was causally related or non-causally associated in another task. The results found that the evaluation of causally related words, as well as causal task cue, engaged a mesolimbic and mesocortical circuitry known to mediate prediction error learning (Corlett et al, 2004;Fenker et al, 2010), suggesting that prediction error processing is involved during the assessments of causality even under conditions when it is not explicitly required to make predictions.…”

“…Overall, these studies indicated that the representation of causal judgment was dissociated from associative judgment in semantic memory, and participants appeared to distinguish the roles of cause and effect or bind the relevant events into the 'cause' and 'effect' roles when verifying the causal relations (Fenker et al, , 2010Satpute et al, 2005). However, little is known about the time course of how stored causal relations are represented and accessed, as well as the representation of causal judgment and associative judgment in semantic memory.…”

“…If the processing of related words was different from unrelated words, larger N400 effect should be elicited by unrelated words than for related words. Furthermore, processing differences between causal judgment and associative judgment should also be found, because they involved different degrees of attentional and executive resources (Fenker et al, 2010;Fugelsang & Dunbar, 2009;Satpute et al, 2005).…”