Academic journal article Memory & Cognition

Processing the Presence, Placement, and Properties of a Distractor in Spatial Language Tasks

Academic journal article Memory & Cognition

Processing the Presence, Placement, and Properties of a Distractor in Spatial Language Tasks

Article excerpt

A common way to describe the location of an object is to spatially relate it to a nearby object For such descriptions, the object being described is referred to as the located object; the object to which it is spatially related is referred to as the reference object. Typically, however, there are many nearby objects (distractors), resulting in the need for selection. We report three experiments that examine the extent to which a distractor in the display is processed during the selection of a reference object. Using acceptability ratings and production measures, we show that the presence and the placement of a distractor have a significant impact on the assessment of the spatial relation between the located and reference objects; there is also evidence that the properties of the distractor are processed, but only under limited conditions. One implication is that the dimension that is most relevant to reference object selection is its spatial relation to the located object, rather than its salience with respect to other objects in the display.

We interact with objects in a variety of ways: We recognize them, we act on them, and we speak about them. Each of these processes requires selecting an intended object from a set of nonintended objects. For example, with respect to recognition, if I want to make a phone call, I need to distinguish my cell phone from the objects surrounding it on my desk. With respect to language, I can convey the location of the cell phone to a listener who may want to use it. For example, I could spatially relate it to another object in the scene, as in The cellphone is behind the book. In this description, the cell phone is the located object, and the book is the reference object. Selection in this case includes not only the located object, but also the reference object, given the many objects surrounding the cell phone. Indeed, one could describe the cell phone's location in different ways, depending on the reference object selected, such as to the left of the monitor, in front of the coffee cup, or at the corner of the desk. The goal of the present article is to more closely investigate conditions in which a reference object must be selected in the presence of an additional distractor. We are specifically interested in the degree to which the presence, placement, and properties of multiple objects are evaluated during reference object selection.

The Role of Salience in Reference Object Selection

It is typically assumed that a reference object is selected on the basis of properties that make it salient relative to other surrounding objects, and therefore easy to find (de Vega, Rodrigo, Ato, Dehn, & Barquero, 2002; Miller & Johnson-Laird, 1976; Talmy, 1983). Such an assumption is consistent with work in communication that argues that referential descriptions that include distinct attributes that make an object stand out in contrast to surrounding objects are particularly helpful in disambiguating the intended referent (Brown-Schmidt, Byron, & Tanenhaus, 2005; Eberhard, Spivey-Knowlton, Sedivy, & Tanenhaus, 1995; Olson, 1970). For example, Olson argued that a speaker would describe the same small round white block as the white one when in the context of a small round black block, and as the round one when in the context of a small square white block.

There are many possible properties that could potentially define salience, including perceptual, conceptual, and spatial features. With respect to perceptual features, Talmy (1983) argued that the spatial description The bicycle is near the house is more acceptable than The house is near the bicycle because the house is larger and more permanently situated, rendering it a more stable and hence more preferred reference object. In addition to size and stability, Talmy argued that reference objects may have greater geometric complexity. In addition, using a corpora of spatial descriptions, de Vega et al. (2002) found that reference objects were more likely to be (1) inanimate, (2) solid in their consistency, and (3) whole entities or mass objects (e. …

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.