Predictive versus diagnostic causal learning: Evidence from an overshadowing paradigm

Causal directionality belongs to one of the most fundamental aspects of cau sality that cannot be reduced to mere covariation. This paper is part of a debate between proponents of associative theories, which claim that learner s are insensitive to the causal status of cues and outcomes, and proponents of causal-model theory, which postulates an interaction of assumptions abo ut causal directionality and learning. Some researchers endorsing the assoc iationist view have argued that evidence for the interaction between cue co mpetition and causal directionality may be restricted to two-phase blocking designs. Furthermore, from the viewpoint of causal-model theory, blocking designs carry the potential problem that the predicted asymmetries of cue c ompetition are partly dependent on asymmetries of retrospective inferences. The present experiments use a one-phase overshadowing paradigm that does n ot allow for retrospective inferences and therefore represents a more unamb iguous test of sensitivity to causal directionality. The results strengthen causal-model theory by clearly demonstrating the influence of causal direc tionality on learning. However, they also provide evidence for boundary con ditions for this effect by highlighting the role of the semantics of the le arning task.