According to current usage, the term frame refers to how individuals perceive view and interpret features of the world. Similarly, researchers in the areas of human decision-making, cognition and communication, use the concept of framing to refer to the way individuals perceive or attach meaning to an issue (Gray, Purdy & Bouwen, 1990; Lewicki et al., 2001). In this respect, framing is unavoidable: as soon as individuals interpret features of their environment, they are selecting to focus on the information available to them (Lewicki et al., 2001). How they interpret this information or what features they focus upon will unquestionably affect the way we subsequently choose to behave. In this sense, framing is a critical component of decision-making and problem-solving (Pruitt & Carnevale, 1993). Conceptualised in this way, framing will clearly govern the way issues are defined and interpreted between two people, as in the case of negotiation. Within the domain of negotiation research, the way in which information is framed has been found to influence the expectations and preferences of negotiators (Tversky & Kahnemann, 1981; Neale & Bazerman, 1991), their interpretations of the process (Donnellon & Gray, 1989), their evaluations of the other party (Thompson & Hastie, 1990), the strategies and tactics they adopt (de Dreu, Carnevale, Emans & Van de Vliert, 1994), and the type and quality of outcomes that they achieve (Bottom & Studt, 1993; Gray, Younglove-Webb & Purdy, 1997; Olekalns, 1991; Pinkley & Northcraft, 1994).
Most negotiation theorists conceptualise frames as mental constructions of events. However, the concept of framing has been employed by different theorists in various ways. Within the broader domain of psychological research, some scholars conceive of frames as conceptualisations of problems or as categories of experience that guide the way in which events are interpreted (Bartlett, 1932); other theorists employ the term more specifically to refer to decision rules that are used in judgement tasks. In the area of negotiation research, two approaches to framing predominate. The first stems from a cognitive heuristics perspective. Drawing on the work of Kahneman and Tversky (1979), this perspective applies the frame construct to understand the breaches of rationality that disputants routinely commit in negotiation-based settings (Neale & Bazerman, 1992). The second approach attempts to understand negotiation behaviour by first categorising frames as they are revealed through negotiator communication and then determining how the use of particular frame types influences the negotiation process and its outcomes (Donnellon & Gray, 1989).
Research on cognitive heuristics uses the term frame to refer to cognitive biases that influence decision-making. This approach explores the way that negotiators shift from rational behaviour toward non-rational decision-making. A bias in the way that information is framed by one individual will affect the way that their behaviour is perceived by the other (de Dreu et al., 1994; Neale & Bazerman, 1985; Neale, Huber & Northcraft, 1987; Tversky & Kahneman, 1981). To elucidate further the way that this type of decision-making influences behaviour, two theories should be considered: Kahneman and Tversky's (1979) well-known prospect theory and behavioural decision theory. Prospect theory defines frames as "a decision maker's conception of the acts, outcomes, and contingencies associated with a particular choice" (Tversky & Kahneman, 1981, p.453). Behavioural decision theory postulates that there are systematic ways in which negotiators form judgements that deviate from optimality or rationality (Neale & Bazerman, 1985). In combining behavioural decision theory and prospect theory, the cognitive heuristics' perspective conceives of frames as conscious constructs that are the perception of biases associated with choice (Putnam & Holmer, 1992). For example, the types of biases evident in decision-making that are associated with perceptions of loss or gain. Using this theory several cognitive processes have been investigated including negotiators' perceptions of loss and gain, their orientation towards risky decision-making, their overconfidence and the way that they reach decisions based on certain reference points, called anchoring (Lewicki et al., 2001). These biases can sometimes be ameliorated once negotiators engage in appropriate analysis of events and the associated decisions. Framing that occurs in this way is commonly conceived of as a conscious cognitive activity. To characterise the frames identified in this area of research, behavioural decision theorists have adopted the term reference frames (Schweitzer & De Church, 2001).
Cognitive heuristics' researchers, who investigate frame-related behaviour, typically regard the negotiation process as static, not dynamic. The presentation of frames is manipulated from the outset to observe the impact upon decision-making behaviour. Frames are essentially treated as inputs to the negotiation and are assumed to remain constant throughout the course of the process. This differs sharply from the categorisation approach to framing reviewed in the following section.
To develop a conceptualisation of framing, social psychologists have borrowed key ideas from the cognitive heuristics approach to frames. Their conceptualisation differs, however, in that they are interested in how negotiators perceive events rather than how negotiators reach decisions based on the adoption of frames. In viewing negotiation as a social decision-making process, social psychologists assume that, during the course of a negotiation, disputants act not only to determine the objects of the dispute, but also to interpret and understand the features of the framework in which they are operating (Mather & Yngvesson, 1981). This interpretation utilises information gained from previous experiences that, in turn, shape perceptions of themselves, the other party, the interaction and the content of the dispute. In interpreting such features, disputants employ certain perceptual orientations or frames that in negotiation research are referred to as "conflict frames".
According to researchers who adopt the frame categorisation perspective, conflict frames can be conceptualised as perceptual lenses that serve to "increase the salience of frame-relevant information and decrease the salience of frame-irrelevant information" (Pinkley & Northcraft, 1994, p.194). Conflict frames are characterised as cognitions and thus they cannot be directly observed. Rather, they are identified and investigated as expressions of these cognitions. In other words, it is the behaviour that is used to infer the presence of conflict frames (Lewicki et al., 2001). Although individuals might be able to describe the frames they have used, the most common type of behaviour analysed when studying conflict frames is the communication that occurs between negotiators (Donnellon & Gray, 1989; Pinkley, 1990). Although conflict frames are typically inferred from the content of communication they are not implied by the substance of the communication; rather, they refer to its orientation. To illustrate this difference, consider the following utterance as it occurred in the context of a negotiation: "What you said just then, actually makes me feel better about signing off on an agreement with you." In this utterance, the negotiator can be argued to be framing her communication, not in terms of the signing of an agreement, but in terms of her feelings toward the other party or in terms of the relationship that has developed between the two parties.
Frame category theorists commonly define conflict frames as cognitions that are expressed through negotiator communication. However, one recently developed theory, communicative framing theory, circumvents the problem of inferring frames from communication by conceptualising frames as the communicative structures themselves (Drake & Donohue, 1996). This theory draws on frame category theory in that it is focussed on exploring meanings and interpretations of conflict by labelling the orientation of negotiator communication. However, this perspective differs from frame category theory in that conflict frames are characterised as the actual communication that occurs between disputants and not as the cognitions that produce the communication. Such a distinction does not lead to differences in the way that research is conducted, but may nonetheless influence the practical implications of findings. For example, if the adoption of a certain type of frame is found to promote a more optimal outcome, communicative framing theory may recommend that a particular set of statements be expressed. In contrast, more cognitively oriented frame theorists may recommend that certain types of cognitions be adopted.
Irrespective of whether frames are defined as cognitions that are expressed through communication or as the communication itself, other negotiation-related constructs may also be expressed through communication. Such features include the strategies and tactics that negotiators use and the goals they pursue. Despite differences in the way that these concepts are defined, there are certain challenges involved in operationalising these variations. These challenges are described in the following discussion.
Whereas conflict frames represent the way an issue is defined at a particular point in time, goals represent an individual's aims or reasons for negotiating. Even though goals are typically developed prior to negotiating, like frames, they can be modified during the course of negotiation (Wilson & Putnam, 1990). As a negotiation proceeds, changes in the way an individual defines a situation, that is, a reframing of information, may alter the types of goals a disputant wants to achieve. Despite this similarity, however, goals and frames can nonetheless be differentiated through the level of intentionality associated with them: Whereas goals are intentions that are formulated and typically reflected upon, conflict frames are inescapably formed perceptions or orientations about the world.
Conflict frames can also be differentiated from strategies. Conflict frames are more indicative of the way that information is perceived and expressed, whereas strategies reflect the way that negotiators have chosen to pursue their goals during a negotiation. The way that information is framed is then used to formulate the negotiator's goals and strategies. Tactics are similar to strategies in that they reflect the method that a negotiator has chosen to attain certain outcomes in the negotiation. Tactics can be differentiated from strategies in terms of their scope: Whereas strategies refer more to the overarching mode of operation that a negotiator chooses to adopt to achieve certain goals in the interaction, tactics are more discrete behaviours that are used to implement the selected strategy. Again, the main difference between frames and these other concepts, however, pertains to the level of intentionality: Strategies, tactics, and goals are typically intentional, and are reflected upon by negotiators, but frames are rarely intentional; they are more contingent upon our cumulative experiences.
Although these differences are important to articulate, the study of the way conflict is framed focuses not upon the extent to which certain activities are intentional, nor upon how experiences govern the development of individual frames, but more upon the way that the framing of information can influence the negotiation process. To assess this influence, researchers in this field have principally used two methods of measurement: one that measures conceptions of conflict before and after negotiation, and the other that focuses on the way that frames are adopted and readopted during the negotiation process.
In employing the frame categorisation approach, Pinkley (1990) analysed negotiators' descriptions of conflict before and after they engaged in negotiation. Using multidimensional scaling techniques, Pinkley uncovered three reliable dimensions of conflict frames to form the basis of a frame typology. These were categorised as cooperate-win, intellectual-task and relational-emotional frame dimensions. Accordingly, the terminology Pinkley used to describe this form of conflict frame was dimensions of "conflict frame".
In interactive settings, when new inputs are experienced, interpretations of these inputs are fed back into the process via communication (Goffman, 1974). Through communication, researchers can infer the interpretations that individuals construct about ongoing events (Carroll & Payne, 1991; Tannen, 1994). Donnellon and Gray (1989) adopted this approach to the measurement of conflict frames, which they integrated with a cognitive conception of frames. Specifically, they treated frames as cognitive structures that are "displayed in, and changed through, the communication which comprises negotiation." (p.2). In analysing negotiator communication as it occurred, these researchers were able to characterise the dynamic way in which conflict frames are mutually presented, incorporated and fed back into the process. This method of analysing framing as it occurs represents the second way in which frames are measured using the frame categorisation approach.
Other concepts that resemble the conflict frame concept include conflict domains (Wehr, 1979) and negotiator focus (Ury, Brett & Goldberg, 1993). Despite differences in the terminology employed, these approaches are aligned in that negotiator communication is used to infer conflict domains or negotiator focus (Donnellon & Gray, 1989; Donohue, Drake & Roberto, 1994; Pinkley, 1990; Pinkley & Northcraft, 1994; Ury et al., 1988; Wehr, 1979). Moreover, as the aim of this approach is to characterise negotiations according to the dynamic focus inherent in negotiators' communication as it occurs, these approaches can be aligned with the frame categorisation approach as employed by Donnellon and Gray (1989).
Even though the cognitive heuristics and the frame categorisation approaches typically share the contention that frames are mental representations of events, their research differs largely in its application. Whereas the cognitive heuristics perspective is interested in how negotiators reach decisions based on their conceptions of events, the frame categorisation approach aims to identify how events are perceived and interpreted (Pinkley, 1990; Putnam & Holmer, 1992). Each approach employs different methodologies to investigate framing. The cognitive heuristics approach treats frames as static cognitions and, accordingly, assesses negotiator behaviour once pre-negotiation information is framed in a particular way. The frame categorisation approach analyses changes in frame use across the course of negotiation. Studies employing the frame categorisation approach either measure frames pre- and post-negotiation (Pinkley & Northcraft, 1994) or they adopt a more direct approach and infer shifts in frames as they occur, through an analysis of negotiator communication (Donnellon & Gray, 1989; Drake & Donohue, 1996; Gray, 1997; Gray, Purdy & Bouwen, 1990). Proponents of the frame categorisation tradition who analyse ongoing communication argue that the cognitive heuristics approach and other approaches to frame categorisation are all limited by their disregard of the principal behavioural component of negotiation, namely negotiator communication (Gray et al., 1990). In treating frames as features that are predetermined by or simply the result of negotiation, the negotiation process is effectively rendered a "black box" (Gray et al., 1990). As the current research focus is upon examining how frame use changes over the course of a negotiation, the content of this black box needs to be explored. Thus, in line with the approaches of Donnellon and Gray (1989) and Drake and Donohue (1996), the current research has used ongoing negotiator communication to explore the phenomenon of conflict frames.