No self-advantage in recognizing photographs of one's own hand: experimental and meta-analytic evidence

Visually recognising one's own body is important both for controlling movement and for one's sense of self. Twenty previous studies asked healthy adults to make rapid recognition judgements about photographs of their own and other peoples' hands. Some of these judgements involved explicit self-recognition: "Is this your hand or another person's?" while others assessed self-recognition implicitly, comparing performance for self and other hands in tasks unrelated to self-other discrimination (e.g., left-versus-right; match-to-sample). We report five experiments with three groups of participants performing left-versus-right (Experiment 1) and self-versus-other discrimination tasks (Experiments 2 to 5). No evidence was found for better performance with self than with other stimuli, but some evidence was found for a self-disadvantage in the explicit task. Manipulating stimulus duration as a proxy for task difficulty revealed strong response biases in the explicit self-recognition task. Rather than discriminating between self and other stimuli, participants seem to treat self-other discrimination tasks as self-detection tasks, raising their criterion and consistently responding 'not me' when the task is difficult. A meta-analysis of 21 studies revealed no overall self-advantage, and suggested a publication bias for reports showing self-advantages in implicit tasks. Although this may appear counter-intuitive, we suggest that there may be no self-advantage in hand recognition.