Midsession reversal learning performance in harbour seals (Phoca vitulina)

Reversal learning (RL) experiments explore cognitive flexibility and decision-making processes. Specifically, RL examines the extent and speed at which individuals adapt their choices when reward contingencies change after the point of reversal. One variation of RL is the midsession reversal learning experiment (MRL), in which the point of reversal occurs midway through a session. In order to solve an MRL task, individuals may employ a win-stay/loose-shift (WS/LS) strategy, changing the behaviour after experiencing a single error after the point of reversal, or utilize a switching cue such as the time elapsed since the onset of the session to solve an MRL task. In this study, we conducted a spatial MRL experiment with two harbour seals to answer the question of whether/how individual learning strategies and temporal information influence decision making and cognitive flexibility in harbour seals. The research underscores individual variations in strategies, which are likely rooted in the seals' experimental pasts. One seal, Moe, with prior RL experience, demonstrated a WS/LS strategy illustrating the impact of feedback on behaviour. Conversely, Bill, naïve to RL tasks, showed anticipatory and perseverative errors close to the reversal point, hinting at the incorporation of temporal information into his decision-making process. Thus, our study showcases cognitive flexibility influenced by learning strategy and possibly temporal information. Although the use of strategies has already been speculated about in previous studies, we now present evidence that temporal information may prompt behavioural adjustments. In the harbour seals' natural habitat, temporal aspects might inform decisions in foraging or orientation/navigation contexts, providing avenues for future research.