Effects of autobiographical remembering in the repetition priming of visual word identification
This thesis is concerned with current debate about alternative 'lexical' and 'episodic' accounts of repetition priming in visual word perception. Like some previous research, the present study investigated effects of 'context congruence' of pre-test and test word presentations. However, modified methods were employed to limit methodoligical problems observed in previous research and to treat the issue of strategic control in repetition priming. The experiments investigated the effect of test orienting tasks which either did or did not require subjects to engage in deliberate remembering of pre-test context. A recognition-memory orienting task was employed to induce deliberate remembering and a letter-judgement orienting task was employed to avert deliberate remembering. Experiment 1 demonstrated a strong effect of orienting tasks in a naming task; repetition priming was reliably greater in the recognition-memory condition. The effect was replicated in a 'restricted' tachistoscopic identification task in Experiment 2, suggesting that the locus of the effect was within processes integral to word identification. Experiment 3 showed that the effect could not be attributed to an inadvertent masking of repetition priming in the letter-judgement conditions, or to be manipulation of subjects' prior knowledge of repetitions. Although these results provide new evidence of episodic memory coding in repetition priming, it can be argued that they do not necessarily imply that all repetition effects depend upon episodic memory coding. The remembering-enhanced repetition effect might reflect the superimposition of an exceptional autobiographical-memory repetition effect upon a normal lexical repetition effect. To test this possibility, Experiments 4 through 7 tested for evidence of dual memory components as a functional dissociation between normal and remembering-enhanced repetition effects. No evidence of a dissociation was found for the following expprimental manipulations: (1) modality of pre-test word presentation, (2) word frequency, (3) subjects' confidence criteria for word naming, (4) 'level of processing' of pre-test words. The discussion considers theoretical and methodological implications, and reviews some related research. The main conclusion suggests that the findings of this thesis are consistent with the assumption of a unitary episodic memory system underlying repetition priming and other phenomena of learning and memory, and that the findings pose some problems for alternative accounts.