Episodic storage and retrieval in visuo-spatial tasks: evidence of indirect retrieval

ABSTRACT

Episodic storage contends that the relevant target and irrelevant distractor features of a display are processed and then stored together within the boundaries of an episode, and are linked in this way. Accordingly, the selective and “direct” retrieval of either a stored target or distractor should result in the “indirect” and unintended retrieval of the other stimulus. In the present investigation, interest focused upon spatial tasks and the “indirect” retrieval of prime distractor processing. The novel findings obtained supported the existence of the “indirect” retrieval of prime distractor processing in visuo-spatial tasks and, to a lesser extent, of the stored prime target event, consistent with identity tasks (i.e. bi-directional episodically-based retrieval). Clearly, more is retrieved than is presented (Hintzman, D. L. (1984). MINERVA 2: A simulation model of human memory. Behavior Research Methods, Instruments, & Computers, 16, 96–101. doi:10.3758/BF03202365), as an episodic storage structure would predict.

KEYWORDS:

• Episodic storage
• error protection
• location tasks
• spatial negative priming

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Contrary to the viewpoint embraced in this study, Neill and Kleinsmith (2016) have concluded that the prime distractor's location, and not its related response, is the source of the delay on ignored-repetition trials, and hence is the cause of the visual spatial negative priming (SNP) effect. We believe that this conclusion does not apply to the visual SNP task as structured here, and as employed in the prior work pointing to the prime distractor's response as the visual SNP locus (e.g. Buckolz et al., 2004, 2012, 2015; Fitzgeorge & Buckolz, 2008; Fitzgeorge et al., 2011; Guy et al., 2006). A proper accounting as to why this is the case is beyond the scope of this footnote. Suffice it to say that one cluster of factors giving rise to the disconnection of the Neill and Kleinsmith study is procedural in nature. The current study, along with past work, has used centrally positioned event locations (i.e. central SNP task) to avoid the confounding engendered when peripheral locations are employed (i.e. “orientation inhibition”, and perhaps more precisely, a “net inhibitory vector” that confounds distractor response inhibition after-effects: Fitzgeorge & Buckolz, 2009; Klein, Christie, & Morris, 2005). Neill and Kleismith used peripheral event locations. Additionally, with visual central SNP tasks, the SNP phenomenon is removed when 100% distractor-free probe trials are used (Buckolz et al., 2002; Guy et al., 2004; Tipper, Brehaut, & Driver, 1990: also see Fitzgeorge et al., 2011; Haworth et al., 2014). Neill and Kleinsmith utilised 100% distractor-free probe trials. Finally, other procedures used by Neill and Kleinsmith that may have mitigated against finding a response locus for visual SNP is that their degree of location-response compatibility was too low; this issue was not addressed in the manuscript, and is difficult to determine on the basis of the Methods description. Moreover, the trials used to calculate a location locus included a response conflict component, not contained in the respective control trials (or in the response locus trials). This might have contributed to, or fully caused, the location locus that Neill and Kleinsmith reported.

2 Other explanations as to how the .25 [probe distractor present]/.75[probe distractor absent] manipulation achieves visual SNP effect removal could be considered if other viable accounts of visual SNP production were available. To be sure, other credible explanations for SNP effects have been advanced, but these hold for tactile (e.g. Wesslein et al., 2016) and auditory (e.g. Mayr et al., 2011) sensory systems, but do not apply to visual SNP (also see Footnote 2). In any event, an explanation as to how the probe distractor manipulation in Exp. 1B causes visual SNP removal is not required for its use in this study. This manipulation was used as part of a “remove SNP” –“restore SNP” two-step strategy; the first step being caused by the probe distractor manipulation. The “restoration” of the SNP effect then allowed us to detect the “direct” retrieval of the stored prime-trial distractor (identity matching prime-probe distractor identities [D→D]).

3 On the notion of irrelevant display aspects: we have previously proposed in this report that a distinguishing feature of an episode is that both relevant and irrelevant aspects of a visual display, possibly along with the required response, are stored together. We conceptualized irrelevant as being those aspects that do not directly contribute to response selection. However, the stored display aspects so defined could be considered relevant if they enhance the “distinctiveness” of the episode, and so improve its retrievability (e.g. see Neill et al., 1992). It is an open question as to whether all aspects stored within an episode need to be deemed relevant as defined in this report, or whether (irrelevant) aspects that do not assist with response selection, or with enhancing episode distinctiveness, are also retained.

Additional information

Funding

This work was supported by a grant from the Natural Sciences and Engineering Research Council of Canada to the second author.