The effect of temporal distance on level of mental construal

Four studies examined the prediction of construal level theory that the more distant future events would be construed in higherlevel, more abstract, and simple terms. Objects were categorized into broader categories when they pertained to distant future situations than to near future situations (Study 1). Positive and negative experiences in the more distant future were expected to be more prototypical—less variable and more extreme (Study 2). More distant future coping experiences were expected to be less variable (Study 3). Preferences for events and activities that were expected in the distant future were organized around simpler structures than preferences for the same events when they were expected in the near future (Study 4). These results support the principle of temporal construal, according to which the more distant future is represented in a more schematic, abstract, and coherent way. 2002 Elsevier Science (USA). All rights reserved. A large amount of empirical research as well as anecdotal evidence suggest that temporal distance from future events systematically changes people’s responses to those events. Thus, research in behavioral economics has found that future outcomes typically undergo discounting at a rate that is steeper than would be justified by purely economic considerations (for reviews, see Frederick, Loewenstein, & O’Donoghue, 2001; Loewenstein & Prelec, 1992). Research on decision making has shown that people often take more risk and feel more confident about the more distant future (Gilovich, Kerr, & Medvec, 1993; Nisan, 1972). Research on delay of gratification and self-control has shown that both humans and animals are better able to delay gratification with respect to distant future outcomes than near future outcomes (Ainslie, 1975; Mischel, 1974; Rachlin, 1995; Trope & Fishbach, 2000). This research has identified important regularities in temporal changes in responses to future events but paid relatively little attention to the cognitive mechanisms that could mediate these effects (but see Mischel, Shoda, & Rodriguez, 1989). The present paper is designed to close this empirical gap: we explore the cognitive mechanism that may underlie the effects of future time perspective on judgment, evaluation, and decision. Our approach to the question of intertemporal evaluation and choice is based on the social cognitive view that people’s responses to social events are based on how they mentally construe those events (Griffin & Ross, 1991; Pennington & Hastie, 1988, 1993; Ross, 1990; Semin & Fiedler, 1988; Semin & Smith, 1999; Smith, 1998; Trafimow & Wyer, 1993; Vallacher & Wegner, 1986, 1987; Wilson & Brekke, 1994). This view would suggest that temporal distance changes people’s responses to future events by changing the way they construe these events. Construal level theory (CLT) specifically proposes that mental construal involves abstraction and that temporal distance is one of the factors that determine the level of abstraction (Liberman & Trope, 1998; Trope & Liberman, 2000). Previous research has already demonstrated that distant future actions are construed more abstractly than near future actions (Liberman & Trope, Study 1). The purpose of the present paper is to extend these findings to objects of construal beyond actions and to levels of construal beyond action hierarchies. According to CLT, people construct more abstract representations (high-level construals) of information pertaining to distant future events and more concrete representations (low-level construals) of information Journal of Experimental Social Psychology 38 (2002) 523–534 www.academicpress.com Journal of Experimental Social Psychology Corresponding author. E-mail address: [email protected] (N. Liberman). 0022-1031/02/$ see front matter 2002 Elsevier Science (USA). All rights reserved. PII: S0022 -1031 (02 )00535-8 pertaining to near future events. For example, in construing the more distant future, people would be more likely to use stereotypes rather than individuating information, generalized scripts rather than more concrete, non-schematic details, trait concept rather than situation-specific goals, and causal explanations rather than unconnected occurrences. As a result, the evaluative implications of the more abstract construals would be more prominent in the more distant future. For example, a distant future behavior is likely to be represented as ‘‘expressing friendliness’’ and thus seem quite positive, but when imagined in the near future the same behavior would be construed as ‘‘dialing a friend’s number’’ and thus seem fairly neutral (Nussbaum, Trope, & Liberman, 2001a). How does abstraction change the meaning of an event? A defining property of concrete representations is that they lend themselves to multiple abstract representations. For example, the concrete action ‘‘waving the hand’’ could be identified more abstractly as threatening or, alternatively, as being friendly (Vallacher &Wegner, 1987), and ‘‘a poodle’’ could be classified as a pet or, alternatively, as a mammal (Rosch & Lloyd, 1978). Usually, some aspects of the focal stimulus are more closely related to one interpretation or categorization (e.g., the hand movement was relatively fast; poodles are friendly), while other aspects are more closely related to the other interpretation or categorization (the person who waved her hand seemed to smile; poodles are warm blooded). Moving from a concrete representation to an abstract representation involves deciding on one of the alternative abstract representations. This, in turn, means omitting the features that are perceived to be less important while retaining those considered more central or important to the construct in question. For example, in replacing ‘‘waving the hand’’ with the more abstract construal ‘‘showing friendliness,’’ the fact that one used one’s hand is omitted (Semin & Fiedler, 1988). Similarly, in representing a poodle as a pet, warm-bloodedness is omitted (Rosch & Lloyd, 1978). Like irrelevant details, details that are inconsistent with the chosen abstract representation are omitted from the representation or assimilated into it. For example, the detail that the hand waving was slightly faster than usual would be omitted or modified once the ‘‘being friendly’’ interpretation is chosen. Because abstract representations necessarily impose one of a few possible alternative interpretations, and because irrelevant or inconsistent details are omitted from the abstract representation or assimilated to it, abstract representations may be expected to be simpler, less ambiguous, and more prototypical than concrete representations (Fiske & Taylor, 1991, p. 98; Smith, 1998). Temporal construal theory thus proposes that the distant future is construed in a more abstract way, and thus construals of the distant future are more schematic—less ambiguous and more simple and coherent than construals of the near future. There are multiple levels of abstractness, as one could construct less inclusive or more inclusive categories of objects (e.g., poodle, dog, and mammal). In featurebased theories of categorization, more inclusive categories have fewer features and therefore are simpler than concrete categories (Rosch & Lloyd, 1978). Actions also form hierarchies of abstractness (e.g., giving money, helping, and being a good person) with each level of abstractness containing less concrete details about the specific type of action performed and the objects it involved (Semin & Fiedler, 1988; Trope, 1989). In the same way, traits form hierarchies (e.g., an excellent guitarist, musical, and talented) such that more abstract traits are less detailed about the behaviors, objects, circumstances, and people it involves (Hampson, John, & Goldberg, 1986). Goal-directed actions form hierarchies too, as goals could be translated into more abstract superodinate goals (Carver & Scheier, 1990, 1999; Miller, Galanter, & Pribram, 1960; Vallacher & Wegner, 1987). In such hierarchies, each action (e.g., studying for an exam) has a superodinate, abstract level, which answers the question of why the action is performed (e.g., doing well) and a subordinate, concrete level, which supplies the details of how the action is to be performed (e.g., reading a textbook). In sum, we suggest, in line with other theories of mental construal, that abstract mental models represent the rich, detailed and possibly ambiguous information contained in a real social event in a relatively simple and coherent mental model. Moreover, it is possible to think about levels of abstractness as a gradual reduction in details and complexity of representations. We further suggest that higher level, more abstract, simple, and coherent representations are formed of events in the more distant future. Our research so far has examined the implications of CLT for evaluation, prediction, and choice (Liberman & Trope, 1998; Nussbaum, Trope, & Liberman, 2001b; Sagristano, Trope, & Liberman, 2001; Trope & Liberman, 2000). These studies demonstrate that the informational and evaluative implications of superordinate, general aspects of future events are more influential in determining responses to distant future then near future events. For example, one series of studies investigated temporal changes in the influence of information about superordinate goals (‘‘why’’ aspects of action) and information about subordinate means for reaching those goals (‘‘how’’ aspects of action) on evaluation and choice (Liberman & Trope, 1998). These studies found that information about ‘‘why’’ aspects of actions was more influential in decisions for the distant future, whereas information about ‘‘how’’ aspects of actions was more influential in decisions for the near future. Another series of studies by Nussbaum et al. (2001b) 524 N. Liberman et al. / Journal of Experimental Social Psychology 38 (2002) 523–534