Academic journal article Perception and Psychophysics

Nothing Compares 2 Views: Change Blindness Can Occur despite Preserved Access to the Changed Information

Academic journal article Perception and Psychophysics

Nothing Compares 2 Views: Change Blindness Can Occur despite Preserved Access to the Changed Information

Article excerpt

Change blindness, the failure to detect visual changes that occur during a disruption, has increasingly been used to infer the nature of internal representations. If every change were detected, detailed representations of the world would have to be stored and accessible. However, because many changes are not detected, visual representations might not be complete, and access to them might be limited. Using change detection to infer the completeness of visual representations requires an understanding of the reasons for change blindness. This article provides empirical support for one such reason: change blindness resulting from the failure to compare retained representations of both the pre- and postchange information. Even when unaware of changes, observers still retained information about both the pre- and postchange objects on the same trial.

Usually, one's phenomenological experience is of a rich and stable visual world, but where does this sense come from? This richness might arise from a detailed internal representation of the world that can be compared or combined with subsequent views (e.g., Marr, 1982; McConkie & Rayner, 1976; Poggio, Torre, & Koch, 1985; Trehub, 1991). Alternatively, it might arise from the absence of such a representation combined with instantaneous access to the outside world when needed (e.g., Gibson, 1966; Grimes, 1996; Hayhoe, 2000; O'Regan, 1992; O'Regan & Noë, 2002; Rensink, 2000; Stroud, 1955). Over the past decade, change detection tasks have been used extensively to infer the quality and detail of internal visual representations. The primary finding from a variety of experiments is that substantial visual changes can go undetected-a phenomenon known as change blindness (for recent reviews, see Hollingworth & Henderson, 2002; Rensink, 2002; Simons, 2000).

In recent years, researchers have explored several causes of change blindness (see Simons, 2000, for a discussion). One explanation suggests that change blindness results from a failure to encode or represent the prechange information (Noë, Pessoa, & Thompson, 2000; O'Regan & Noë, 2002). Unless observers encoded and retained a representation of the initial display, they could not detect a change. Another possibility is that change blindness occurs when adequately encoded information about the initial display is disrupted, overwritten, or forgotten (Beck & Levin, 2003; Becker, Pashler, & Antis, 2000; Brawn, Snowden, & Wolfe, 1999; Irwin, 1992; Levin, Simons, Angelone, & Chabris, 2002; Pashler, 1988; Phillips, 1974; Rensink, O'Regan, & Clark, 1997; Silverman & Mack, 2001; Taller, 2001; Wolfe, 1999). A third alternative is that change blindness results not from the absence or inadequacy of a representation, but from a failed comparison of a prechange representation to the postchange information (Angelone, Levin, & Simons, 2003; Hollingworth & Henderson, 2002; Ryan & Cohen, 2004; Scott-Brown, Baker, & Orbach, 2000; Shore & Klein, 2000; Simons, 2000; Simons, Chabris, Schnur, & Levin, 2002).

Determining which of these mechanisms contribute to change blindness is central to understanding the completeness of internal representations. If change blindness results entirely from the absence of representations, representations of the visual world might well be sparse, with little or no information retained internally. This view is consistent with the idea that the world itself might act as an "outside memory," with no need to store visual details internally (O'Regan, 1992; O'Regan & Noë, 2002). Alternatively, if change blindness results solely from overwriting of the prechange representation, representations of the visual world might be detailed, but fleeting and fragile. Lastly, if change blindness results from failed comparisons, then representations of the visual world might be detailed and relatively stable even if they are not always readily accessible or cannot be aligned between pre- and postchange views. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.