An effect of context on whether memory for initial position exhibits a Fröhlich effect or an onset repulsion effect

Abstract

Memory for the initial and final positions of moving targets was examined. When targets appeared adjacent to the boundary of a larger enclosing window, memory for initial position exhibited a Fröhlich effect (i.e., a displacement forward), and when distance of initial position from the boundary increased, memory for initial position exhibited a smaller Fröhlich effect or an onset repulsion effect (i.e., a displacement backward). When targets vanished adjacent to the boundary of a larger enclosing window, memory for final position was displaced backward, and when distance of final position from the boundary increased, memory for final position did not exhibit significant displacement. These patterns differed from previously reported displacements of initial and final positions of targets presented on a blank background. Possible influences of attention and extrapolation of trajectory on whether memory for initial position exhibits a Fröhlich effect or an onset repulsion effect and on backward displacement in memory for final position are discussed.

Acknowledgments

The authors thank Jane Montgomery for assistance in data collection and Dirk Kerzel and an anonymous reviewer for comments on a previous draft of the manuscript.

Portions of these data were presented at the Second Annual Meeting of the Vision Sciences Society (Sarasota, Florida, May 2002) and at the European Conference on Visual Perception (Glasgow, Scotland, August 2002).

Notes

Michael Motes is now at Department of Psychology, Rutgers University-Newark, Newark, NJ 07102, USA.

There are two studies that initially appear to be exceptions to this general pattern, but which upon closer examination may not be exceptions. Müsseler and Aschersleben (1998) reported a robust Fröhlich effect for horizontally moving targets presented on a blank background or moving between background scales (i.e., rulers). However, the target always appeared at either one of two highly predictable locations, and when this trial context was eliminated, the Fröhlich effect was also eliminated (see Müsseler & Kerzel, 2004). It may be that the predictability of target location provided a “functional window” within the display, which influenced remembered position much as did a real window. Thornton (2002) reported a robust onset repulsion effect for targets that moved across a grid pattern. However, this grid pattern covered the extent of the background and so may have acted more like a textured background rather than as a spatial context (i.e., rather than as a nontarget stimulus object) per se.

Representational momentum for some types of stimuli may be influenced by whether the observer fixates a single location some distance from the target or is allowed to visually track the target, and this has led some researchers to suggest that displacement may result from pursuit eye movements and a bias toward the fovea (e.g., Kerzel, 2000). However, fixating a stationary point some distance away from a moving target that an observer has to respond to is not a useful strategy in everyday contexts, and it may be that the disruption of displacement with fixation in some experiments may result from a disruption in the flow of information regarding eye movement activity that is normally provided to higher order mechanisms that produce displacement rather than to any casual role of eye movements per se. Numerous studies suggest a role for higher order processes in displacement (for reviews, see Hubbard, 1995, 2004), and, coupled with the failure of an eye movement explanation to account for representational momentum in memory for “frozen action” photographs (e.g., as in Freyd, 1983; Futterweit & Beilin, 1994) and in memory for stimuli undergoing implied motion (e.g., as in Freyd & Finke, 1984), it is unlikely that eye movements are the primary causal mechanism for representational momentum and related displacements. Given this, a more useful methodology in the current experiments would allow observers to use whatever strategy they would normally use in tasks involving spatial memory.

Displacement may be measured along different axes (e.g., previous studies distinguished between displacement along the axis of motion, M displacement, and displacement along the axis orthogonal to motion, O displacement), and the “M” specifies displacement along the axis of motion. Even though no other displacements are of interest in the current study, the “M” qualifier is retained in order to be consistent with previous practice.

Such an account might seem inconsistent with the robust Fröhlich effect that Müsseler and Aschersleben (1998) reported for horizontally moving targets. However, in Müsseler and Aschersleben's study one endpoint of the trajectory was always 5.5–6.5 deg from the central fixation, and the distance travelled by the target was always 5.5 deg, and, consistent with the speculation earlier, the use of a constant trajectory length might have imposed a “functional window” on the display, which then influenced attention and remembered position much as a real window did in Experiments 1, 2, and 3. The notion of such a “functional window” may also be consistent with the observation of Kerzel and Müsseler (2002) that memory for the orientation of a rotating stimulus did not exhibit a Fröhlich effect unless context was present on both sides of that target. In contrast, trajectory lengths in Hubbard and Motes (2002) and in Thornton (2002) were more variable, and so observers would have been less able to impose such a functional window because the trajectory length in any given trial was much less predictable. Such an account is consistent with Müsseler and Kerzel's (2004) findings that Fröhlich effects in the Müsseler and Aschersleben paradigm did not occur if targets appeared at unpredictable locations.