Symmetrical ‘Super Learning’: Enhancing learning using a bidirectional probabilistic outcome

In a learning scenario, with multiple positive or negative (excitatory or inhibitory) predictive cues of a single outcome, cues interfere with or enhance each other during the acquisition process (e.g., Baker et al., 1993). Previous experiments have primarily focussed on cues that signal the presence or absence of binary outcomes. This introduces a motivational and saliency asymmetry between excitatory and inhibitory learning. Here we asked whether learning about both generative (incremental positive outcome) and preventative (incremental negative outcome) causal cues show similar enhancement effects. In three experiments with humans using predictive learning tasks, participants (N = 133) were exposed probabilistic predictive cues for two opposite polarity events. Generative cues caused an increased outcome, while preventative cues decreased the outcome. Learning was assessed via explicit judgments of the relation between the cue and the outcome. We report evidence for symmetrical learning and enhanced learning for both generative and preventative cues. The results are discussed in relation to super learning, an effect derived from theories of competitive learning based on error correction, and from theories of contrasting probability estimates.