Four Script Studies: What Have We Learned

Paul H. Schurr, SUNY Albany [Associate Professor, School of Business, 1400 Washington Ave., Albany, NY 12222]
ABSTRACT - This paper describes four studies designed to examine the influence of scripts on attitudes and judgment. There are two objectives for this discussion. The first objective is to tie together the findings of f our script studies. The second objective is to identify conceptual issues that have arisen from this program of study.
[ to cite ]:
Paul H. Schurr (1986) ,"Four Script Studies: What Have We Learned", in NA - Advances in Consumer Research Volume 13, eds. Richard J. Lutz, Provo, UT : Association for Consumer Research, Pages: 498-503.

Advances in Consumer Research Volume 13, 1986      Pages 498-503

FOUR SCRIPT STUDIES: WHAT HAVE WE LEARNED

Paul H. Schurr, SUNY Albany [Associate Professor, School of Business, 1400 Washington Ave., Albany, NY 12222]

ABSTRACT -

This paper describes four studies designed to examine the influence of scripts on attitudes and judgment. There are two objectives for this discussion. The first objective is to tie together the findings of f our script studies. The second objective is to identify conceptual issues that have arisen from this program of study.

INTRODUCTION

One advantage of a programmatic series of studies is that they offer a chance to evaluate multiple facets of a theory or conceptual model. Before discussing what we have learned from four script studies, the theory that prompted these studies will be discussed. Then, an overview of the conceptual findings of four studies will be presented.

A CONCEPTUAL MODEL

The conceptual model encompasses four main components. The first two components may be viewed as input components: information scored in an individual's memory and information from an organizational (purchasing) task environment. The third component in the model controls the informational input. It is a collection of sense-making psychological mechanisms - knowledge structures or schemata - that combine an individual's stored knowledge with information from an organized task environment (Weick 1979; Calder and Schurr 1981). The fourth component is the output. In this case we are interested in attitudes and judgment as the output of the sense-making schemata.

Of information input, information control, and output components of the conceptual model, the information control mechanism is perhaps the least understood. First, we will discuss schemata and, particularly, scripts as control mechanisms. Then, information input will be discussed, with special attention to the task environment as a trigger for schematic processing.

Schemata as Control Mechanisms

A schema is defined as "an abridged, generalized, corrigible organization of experience that serves as an initial frame of reference for action and perception" (Weick 1979, p. 50). A schema assumes the operation of four central processes: selection of all incoming stimuli for mindful representation, abstraction of the meaning of perceived information, interpretation using prior knowledge to aid comprehension, and integration for storage in memory (Alba and Hasher 1983).

A schema controls information selection by directing attention to a limited portion of stored knowledge and by causing selective perception of information in the environment. By organizing information, a schema gives meaning to stimuli and enables interpretation and comprehension. Sense-making itself seems related to a network of thought connections, which makes the notion of episodic scripts a good starting point for research into information control mechanisms.

Scripts

A script is an event-centered schema in which knowledge is organized like a causal model around expectations concerning a sequence of events or actions (Abelson 1976). Scripts are of particular interest because scripts link the individual with an organizational context and role (Calder and Schurr 1981).

Further, according to Abelson (1981), scripts capture most of the conceptual elements of other types of schemata, yet permit more focused analysis and experimentation because scripts are simple and well-structured. A script is a mental picture - "picture plus caption" in Abelson's terms (1976, p. 34) - representing the action sequences, participants, and physical objects found in a situation. Information is organized around one or more scripts.

The activation of scripts by an individual depends on three general conditions that must be satisfied for scripted processing to occur (Abelson 1981): the individual must have a cognitive representation of a script, an evoking context must be present, and the script must be physically or mentally entered.

Scripts are evoked by a context. Therefore, because a task environment acts as a trigger for scripts, a task environment at least in part determines what information stored in memory will be activated and what environmental information will be selectively perceived. Thus, a script is a significant way for past experience and organizational task environments to influence individuals (Weick 1979; Calder and Schurr 1981; Axelrod 1976).

Input: Information from the Task Environment and Memory

Essential to the process leading to attitudes and judgment are the informational inputs: information from the task environment and from memory.

The task environment. The organizational task environment is defined as the context, structure, and content of a decision problem and decision-making situation (Newell and Simon 1972). It serves two key functions. First, prominent cues in a context activate the scripts that will in turn guide further information processing. Second, as a context for a role occupant's thought processing, the task environment provides information relevant to role behavior and problem solving through the process of perception.

The role of memory. Like the task environment, an individual's memory contributes to information processing in two different ways. First, scripts are stored in memory and oust be activated. Second, memory provides basic units of information - thoughts and beliefs - as input to active processing. That is, individuals have in their memory information obtained from actual and vicarious experience, more information than could ever be processed efficiently for a given problem (Newell and Simon 1972). Part of this reservoir of information is tappet by a script.

In summary, attitudes and judgment are a function of three elements information provided by a task environment, information activated in memory, and information processing guided by sense-making knowledge structures. Knowledge structures, such as scripts, are stored in memory, but activated by the organizational purchase context. Our simple conceptual model suggests knowledge structures provide the basic ingredients for attitudes and judgment (cf., Cialdini, Petty, and Cacioppo 1981; Greenwald 1968; also see Calder and Schurr 1981). Thus, the following research focuses on demonstrating that an understanding of purchasing scripts can result in accurate predictions concerning buyer attitudes and judgment.

A SUMMARY OF FOUR STUDIES

Our research program called for sequential studies that allowed the researcher to learn at each step in the research program. Four studies in our research program are discussed here. The goal was to test the causal relationship between scripts and purchase evaluations in an organizational purchasing context. The approach was to manipulate aspects of the purchasing task environment and examine hypothesized effects on attitudes and judgment. To evaluate script-information interactions, information was also manipulated in two of the four studies.

A brief summary of the four studies is provided, along with a reference to where more detailed information can be obtained about the study. Then the studies are critiqued from the standpoint of substantive findings and conceptual issues.

Study 1

Background and Hypotheses (see Schurr 1980 for details). The purpose of the first study was to challenge the assumption that the relationship between a purchase decision problem and attitudes is relatively fixed. Our rival hypothesis was that an attitude changes with the perspective (schema) in mind at the time.

In particular, consider boundary role theory (Katz and Kahn 1966; Goffman 1969; Adams 1976; Vidmar 1971). A boundary role person, such as an organizational buyer, performs two representational roles. On one hand, an organizational buyer represents a seller's viewpoint to individuals within the organization, such as users. On the other hand, an organizational buyer also represents the viewpoint of internal constituencies, such as users, to sellers' representatives. Thus, organizational buyers potentially bring two different perspectives to the analysis of every purchase decision problem.

Consider, for example, a situation where there is a disagreement between a user and a supplier of a product. (All four studies focus on purchase problem evaluations stemming from user-supplier disagreements.) Say, for instance, the user claims the competition is promising a better delivery schedule, but the current supplier claims a better delivery schedule is not realistic. As a boundary role person, an organizational buyer must convey the views of the user to the supplier's representative. In seeking a resolution of a purchase-related disagreement, the reverse also holds. That is, the purchasing agent must represent the supplier's view to insiders (i.e., the user) so they will be able to develop realistic expectations about how to resolve the disagreement.

Notice how this example based on boundary role theory lends itself to Abelson's (1976) idea of a script. The basic script concerns the event sequence leading up to and including a meeting between the purchasing agent and one of the disagreeing parties. In the meet-supplier track of the script the participants are a purchasing agent and an outsider (supplier's representative). In the meet-user track of the script the participants are a purchasing agent and an insider (user). We found meetings of this kind to be quite familiar to purchasing agents. The evoking context is provided by the problem (user-supplier disagreement) itself and an anticipated meeting with the user or supplier to discuss the disagreement.

We hypothesized that the meet-supplier induction would cause a purchasing agent to adopt an insider's perspective when expressing attitudes or judgments; the subject would be less favorable toward the supplier's position and more favorable toward the user's position. Essentially, the purchasing agent will organize thoughts to represent the user's viewpoint. By comparison, the meet-user induction would have the reverse effect and cause subjects to be more favorable toward the supplier's position and less favorable toward the user's position in a disagreement.

Method. Twenty-eight organizational buyers rated four items in an attitude scale (Cronbach's alpha = .96) for each of nine different user-supplier disagreements. The between-subject manipulation was simply to interchange the terms "user" and "supplier" when giving instructions to subjects regarding problem evaluations: "You are preparing to meet with the user (supplier) of a product to discuss a disagreement between the user and the supplier...

Results. The main effect was significant at the .001 level (F(1,26) = 12.7). As hypothesized, the meet-supplier script caused a more favorable attitude toward the user's position in the disagreements, as compared to the meet-user script.

Critique. As a first step in the pursuit of attitude-script linkages - indeed, a baby step - the study yielded interesting results. In particular, subjects did not have a patternless reaction to the nine stimulus problems (the null hypothesis). Rather, the meet-user and meet-supplier inductions differentiated the responses. This result suggests something in the "black box" of the human information processors was "switched on" (evoked) in one condition, and something different was evoked in the other condition. Unfortunately, Study l offers little supporting evidence that a script was evoked or activated by the inductions. A remaining task is to demonstrate that scripts were actually involved in the effects.

Study 2

Motivation (see Schurr 1980 for details). Study 2 attempted to enhance the evidence that scripts caused attitude differences in Study l. Thus, one tactic was to carefully design a study in which all the ingredients to script activation would be present. In particular, action sequences, entry conditions, specific roles, and props (physical objects) were included in pictorial stimuli. A second tactic was to manipulate informational inputs so that the interaction of scripts and information processing could be examined. An interaction would suggest that the black box mechanism was handling information differently in different contexts - a theoretical effect of a script or schema. A third tactic was to utilize dependent measures to assess whether action sequences were mindfully represented.

Background and Hypotheses. Study 2 again pursued a substantive issue prompted by boundary role theory. The aim was to utilize role partner relationships to activate script processing. According to Adams (1976, p. 1176), boundary role persons, such as organizational buyers, operate at a distance from the organization and in closer proximity to the outside world. This physical and psychological distance generates suspicion among a buyer's role partners inside the organization and makes a boundary role person sensitive to the reactions of insider role partners, such as a product user. The more exposed a boundary role person is to monitoring by an inside role partner, the more a boundary role person tries to avoid the appearance of compromising the insider's best interests

Following work by Organ (1970), it was hypothesized that a monitored-buyer script (a user is present in a purchasing agent/sales rep meeting) would cause a less favorable evaluation of a supplier's position in a user/supplier disagreement. In an unmonitored script (the user is not present) it was hypothesized that the subjects would evaluate a supplier's position more favorably in relative

A script by information interaction was conceptually motivated as follows. First, Calder, Insko, and Yandell (1974) demonstrated that adding arguments favoring a particular viewpoint increases the persuasiveness of that viewpoint, at least up to the point where the number of arguments swamps short term memory. Second, scripts can increase or decrease the perceived persuasiveness of a communication by influencing meaning, stimulating counter arguments, and affecting attention (Taylor and Crocker 1980). It was hypothesized that the monitored-buyer script would neutralize the persuasiveness of a salesperson's arguments as compared to the unmonitored-buyer script, because the former script would stimulate counter arguments.

Method. Data from forty purchasing personnel in a large company in Chicago were utilized. Monitored/unmonitored buyer scripts were crossed with two levels of supplier's arguments (3 and 6) in a between-subject, two by two factorial design.

Scripts were induced using three small pictures depicting arrival (an entry condition) and meeting actions. Pictures were accompanied by captions identifying participants in the meetings. The user was depicted only in the monitored-buyer version, The objective was to convey realistic but minimal cues consistent with these scripts.

A user-supplier disagreement problem from Study 1 was adapted for use in Study 2. Either three or six supplier arguments (ant three user arguments) accompanied the inductions as presented to the purchasing personnel.

Twenty rating scales items were employed, including five measures designed to assess mentally represented actions prompted by the inductions.

Results. One scale (three items, alpha = .77) indicated the monitored-buyer script caused less favorable ratings of the sales representative's position in the user-supplier disagreement than did the unmonitored-buyer script as predicted (F(1, 36) = 12.1, p < .001).

There was a script by number of arguments interaction, but only for one item that measured how good or poor the user's arguments were perceived by the purchasing agents. The scripts actually reversed the intuitive effect of the number of arguments. That is, monitored buyers rated the user's portion more favorably when there were six, not three, supplier arguments (F(1, 36) = 16.9, p < .001); there was no difference "between three and six arguments for unmonitored buyers.

Other aspects of the results are less conclusive. The buyer's overall attitude toward the supplier was not influenced by the manipulations. Further, there was no script by number of arguments interaction for the primary scales, although a main effect for the number of arguments was significant in the predicted direction (F(1, 36) = 5.6, p < .02).

Finally, the five measures of script-prompted actions yielded little evidence that generic action sequences were brought to mind by the inductions.

Critique. The results indicate the relationship between a boundary role person and his or her role partners is a determining factor when attempting to predict the attitudes and judgment of a boundary role person, such as a purchasing agent.

With respect to scripts, the results are only slightly encouraging. The presence of a script by number of arguments interaction for one item suggests additional research is warranted, but the single-item effect must be recognized as a possible result of chance, since the pattern of results was inconsistent. The main script effect on the three item scale provides more cause for optimism.

Our major conclusion is that the script inductions caused different purchase problem evaluations. Without a convincing interaction effect and without confirming evidence that action sequences were brought to mind by the script inductions, support for script theory is less than hoped for in Study 2.

Study 3

Motivation. The results of the first two studies could be explained by role theory. In fact, since the manipulations simply involved roles, role theory is perhaps the best explanation for the effects. (Nevertheless, script theory explains why simple role manipulations in the context of stereotypical organizational events prompted significant differences in purchase problem evaluations.)

The next two studies focus less on the role aspect of scripts and more on subtle manipulations of the props physical features - present in a stereotypical action sequence. This strategy further examined the notion that the effects in the first two studies were the result of script, rather than role, manipulations.

Background and Hypotheses (see Schurr and Calder 1986 for details). Study 3 provides an interesting connection between the restaurant script, which has received much attention thanks to Abelson's writings, and the every day business lunch. The published report of this study (cited above), in fact, focuses on substantive issues related to buyer-seller interactions at business lunches. Script theory motivates the method used (not the hypotheses) in the published report. In contrast, the discussion here will center on how this study relates to script theory.

Because script research has devoted considerable attention to the notion of stereotypical restaurant scripts (e.g., Abelson 1981), the restaurant script is a firmly anchored starting point. Briefly, script theorists regard the restaurant script as a commonly experienced event sequence around which we organize information. Most people have in their memories a script for traveling to a restaurant, being seated, having a drink, ordering from a menu, being served the first course, and so on.

In organizational buying, an equally common script is the business lunch script (see Schurr and Calder 1986). The business lunch script in Study 3 had a purchasing agent and a supplier's sales representative arriving at a restaurant to discuss, as in the previous studies, a user-supplier disagreement. We are interested in two versions. One is a regular business lunch meeting st an ordinary restaurant. The other version is a meeting at a fancy restaurant. Notice we are primarily interested in the setting, defined by the ordinary or elaborate "props" the restaurant decor.

Based on previous studies of business lunch meetings (Dempsey, Bushman, and Plank 1980; (Halvorson and Rudelius 1977) and Webster and Wind's conceptualization of the buying process (1972), we reasoned that as task environments, the restaurant settings would influence a buyer's interpretation of events by evoking knowledge of specific norms, shared ideals, and positive or negative thoughts that contribute to attitudes and judgment.

The norm for regular meetings at ordinary restaurants is business-as-usual: attention is given to problem-solving and relationship development. Regular business lunch meetings include an element of mutual persuasion, but the motives of the participants are not questioned for the most part.

Fancy restaurants in the context of addressing a disagreement are less normal. While nearly everyone prefers a fancy restaurant to an ordinary restaurant for celebrations and social occasions, when business must be conducted, a fancy restaurant to some extent creates the impression that a buyer's objectivity is impaired.

In our hypotheses we compared the effect of a regular meeting with that of a fancy restaurant meeting. We hypothesized that a regular restaurant meeting would influence an industrial buyer to: (H1) evaluate a sales representative's arguments more positively and a user's arguments more negatively, (H2) judge that the user must yield relative to the sales representative in the buyer seller disagreement, and (H3) give a more positive overall supplier evaluation.

Method. Data was collected from 211 purchasing personnel in the Purchasing Management Association of Chicago by means of a mail survey.

Two scripts and three buyer-seller disagreement problems were evaluated in a six cell between-subjects design. The restaurant meeting script inductions utilized the picture-plus-caption approach of Study 2. Three scenes were depicted: (1) buyer and seller arrive at the door of a restaurant, (2) buyer and seller are shown to a table, and (3) buyer and seller engage in discussion at the table. All three scenes appeared in both restaurant meeting inductions, but the physical appearance of the surroundings - the props in Abelson's terms - were manipulated. The regular restaurant scenes showed unpretentious surroundings and decor, while the fancy restaurant scenes displayed more elegance.

The three disagreement problems were developed from the earlier studies, and four categories of dependent measures were collected.

Results. All three hypotheses were supported at the .025 level or better. Further, four measures designed to capture perceptions of the meeting scripts gave some indication that the respondents were thinking along lines consistent with a script interpretation of the effects (see Schurr and Calder 1986).

Critique. There are several ways to assess the results. First, the substantive finding that norms related to the task environment - in this case the ordinary or fancy restaurant - in part determine attitudes and judgment agrees with theories of organizational buying (e.g., Webster and Wind 1972).

Second, Study 3 demonstrates that the use of script-related inductions allows researchers to explore questions that might otherwise be intractable. In this case, examination of meeting place effects on attitudes was made possible by the script inductions.

Third, Study 3 addresses at least one of the limitations of the first two studies. Compared to the role manipulations in the first two studies, the Study 3 manipulations concerned only the physical aspects of the restaurant settings (nevertheless, role theory was needed to develop the hypotheses). Cognitive reactions to these inductions could not easily be explained without considering that the meeting setting prompted some sort of knowledge retrieval and information organization. Because restaurant scripts have been shown to organize event-related knowledge (Abelson 1981), it is reasonable to suggest that restaurant meeting scripts would, indeed, organize information related to a business lunch meeting task environment.

Fourth, in Study 3 there is more assurance that the subjects were likely to have business meeting scripts stored in memory. Subjects were required to meet two criteria: the respondent must have had experience meeting with supplier's representatives at business lunches, and their primary area of responsibility had to be purchasing.

Study 4

Motivation (see Schurr 1985). Two aspects of Study 3 warranted additional attention. First, it was argued in Study 3 that the meeting script effect was directly related to norms of the task environment associated with the purchasing agent's role and relationship to role partners. This reasoning suggests different roles have different scripts; a seller and buyer should react differently to different stimuli. Second, Study 3 did not examine script effects on information.

Background and Hypotheses. A role by script interaction was predicted. Buyers in Study 4 were expected to react to the meeting scripts as before, with the fancy restaurant meeting resulting in a less favorable attitude and judgment with respect to the seller. By comparison, no script effect was expected for sales representatives.

Study 4 also tested information effects by manipulating subjective or objective information content in the sales representative's response to the user's concerns in the user-supplier disagreement. Compared to subjective information, objective information has been found to cause a more positive response (Edell and Staelin 1983; Holbrook 1978).

An interaction of interest would be a script-induced neutralization of information effects caused by the organization and meaning given to information. We tested the notion that a fancy restaurant meeting would give increased (negative) importance to subjective information.

Method. Data was collected from 248 subscribers to a leading Chicago business letter publication. All of the subjects had direct experience in restaurant meetings with their counterparts. Purchasing and sales people were differentiated based on their responses to a series of questions, such as job title and job responsibilities.

Three independent variables (role, script, information) with two levels each were crossed in an eight cell design. The script and role variables have been described. The information inductions are reported in Schurr (1985). Dependent measures were drawn from the previous studies in this program.

Results. The hypothesized role by script interaction was supported by a two item scale capturing an evaluation of the sale representative. Subjects who were identified as being in a purchasing role in their organizations reacted as did subjects in Study 3: they rated the seller's position in a disagreement less favorably in the fancy restaurant meeting context. Subjects identified as being in a selling role, on the other hand, expressed a positive evaluation of the sales representative, regardless of the script condition. This interaction is significant at the .05 level.

The only script by information interaction was indicated by a single item that concerned a judgment of whether the buyer or seller in the disagreement should make concessions in the disagreement. The concession item indicated that subjectively embellished information neutralized the script effects, while objectively embellished information did not affect the script effects.

Critique. Study 4 did not close the door on script research by any means. On one hand, the role by script interaction supports our conceptualization of scripts, but this interaction was indicated by but one scale; the effect was not pervasive. On the other hand, the information by script interaction was inconsistent with our conceptualization of scripts, but this contradictory evidence was indicated by only a single item.

SUMMARY AND CONCLUSIONS

What We Learned

The results of any single study in this program does not provide sufficient evidence in support of script theory. Yet the accumulated evidence is consistent with script theory. We propose that the following propositions warrant additional consideration, based on the encouraging pattern of results in our research program.

Proposition 1: Scripts affect attitudes and judgment by evoking an organized body of knowledge.

Two kinds of evidence are available. First, previous research has established that people store in memory information related to action sequences (Abelson 1981; Bower, Black, and Turner 1979; Leigh and Rethans 1984). Second, our research program has found effects suggesting a direct relationship between scripts and attitudes and judgment.

Proposition 2. Scripts are role dependent.

Again there appears to be two forms of evidence. First, boundary role theory has served as the conceptual foundation for all four studies. Support for the boundary role hypotheses has been quite consistent. Second, Study 4 demonstrated that individuals in different roles react to scripts differently.

Proposition 3. Scripts capture important aspects of a task environment which in turn frames a buyer's reactions to problems in buyer-seller exchange relations.

All four studies demonstrated that purchasing agents evidence different assessments of user-supplier disagreements, depending on the task environment script in mind at the time

Unsolved Conceptual Issues

Based on these four studies (and the reactions of many who have read reports of these studies), the following issues need attention.

Issue 1: How does script processing relate to information processing explained in terms of cognitive response theory?

A reasonable critique of script research is that many of the effects can be explained in a cognitive response framework. Does script theory represent a more complicated way to explain simple cognitive response? noes script theory explain mental information control processes better than cognitive response theory?

Issue 2: How do we know that scripts, not informational cues-in the inductions, prompted the effects on attitudes and judgment?

Script theory claims that scripts organize a body of knowledge related to attitudes and judgment. A rival hypothesis is that script inductions of the kind used in the above studies simply provide pieces of information that cause the results. If so, an organized body of knowledge - information connected to other information is irrelevant to effects on attitudes and Judgment. This explanation resembles a stimulus-response interpretation.

Conclusion

We believe the conceptual framework offered in this paper provides a reasonable starting point for research. Effort must be directed toward pitting competing theories, such as script theory and cognitive response theory, against one another. Although we cannot disprove these theories, we can demonstrate which theories explain causal relationships more simply and completely.

Finally, based on formal and informal debriefings and discussions with subjects in our studies, we think script-based inductions offer a powerful tool for studying substantive issues that otherwise might not receive attention (e.g., Schurr and Calder 1986). Script-based inductions, with a focus on entry conditions, event sequences, and roles in a task environment, apparently evoke a frame of mind that reflects actual situations.

REFERENCES

Abelson, Robert P. (1976), "Script Processing in Attitude Formation and Decision Making", in Cognition and Social Behavior," eds. John S. Carroll and John W. Payne, Hillsdale. NJ: Lawrence Erlbaum. 33-45.

Abelson, Robert P. (1981) "Psychological Status of the Script Concept," American Psychologist, 36, 715-729.

Adams, J. Stacy (1976), "The Structure and Dynamics of Behavior in Organizational Boundary Roles," in Handbook of Organizational and Industrial Psychology, ed. Marvin Dunette, Chicago: Rand McNally, 1175-1199.

Alba, Joseph W., and Lynn Hasher (1983), "Is Memory Schematic?" Psychological Bulletin, 93, 203-231.

Axelrod, Robert (1973), "Schema Theory: An Information Processing Model of Perception and Cognition," American Political Science Review, 67, 1248-1266.

Bower, Gordon B., John B. Black, and Terrence J. Turner (1979), "Scripts in Memory for Text," Cognitive Psychology, 11, 177-220.

Calder, Bobby J., and Paul H. Schurr (1981), "Attitudinal Processes in Organizations," in Research in Organizational Behavior, Vol. 3, eds. L.L. Cummings and Barry H. Staw, Greenwich, CT: Jai Press, 283-302.

Calder, Bobby J., Chester A. Insko, and Ben Yandell (1974), "The Relations of Cognitive and Memorial Processes to Persuasion in a Simulated Jury Trial," Journal of Applied Social Psychology, 4, 62-93.

Cialdini, Robert B., Richard E. Petty, and John T. Cacioppo (1981), "Attitude and Attitude Change," Annual Review of Psychology, 32, 357-404.

Dempsey, William F., Anthony Bushman, and Richard E. Plank (1980), "Personal Inducements of Industrial Buyers," Industrial Marketing Management, 9 (October), 281-290

Edell, Julie A., and Richard Staelin (1983), "The Information Processing of Pictures in Print Advertising", Journal of Consumer Research, 10 (June), 45-61.

Goffman, E. (1969), Strategic Interaction, Philadelphia: University of Pennsylvania Press.

Greenwald, A.G. (1968), "Cognitive Learning. Cognitive Response to Persuasion, and Attitude Change," in Psychological Foundations of Attitudes," eds. A.G. Greenwald, T.C. Brock. and T.M. Ostrom, NY: Academic Press 147-170.

Halvorson, Paul J., and William Rudelius (1977), "Is There a Free Lunch?" Journal of Marketing, 41 (January), 44-9.

Holbrook, Morris B. (1978), "Beyond Attitude Structure: Toward the Informational Determinants of Attitude," Journal of Marketing Research, 15, 546-556.

Katz, Daniel, and Robert Kahn (1966), The Social Psychology of Organizations, NY: Wiley.

Leigh, Thomas W., and Arno J. Rethans (1984), "A Script-Theoretic Analysis of Industrial Purchasing Behavior," Journal of Marketing, 48 (Fall) 22-32.

Newell, Allen, and Herbert A. Simon (1972), Human Problem Solving, Englewood Cliffs, NJ: Prentice-Hall.

Organ, Dennis W. (1971), "Linking Pins Between Organizations and Environment," Business Horizons, 14, 73-80.

Schurr, Paul H. (1980), SA Cognitive Role Theory of Organizational Buying," unpublished dissertation, Northwestern University. Chicago.

Schurr, Paul H. (1985), "Role, Situation, and Information Effects On Organizational Buyers and Sellers," Working Paper. SUNY Albany.

Schurr, Paul H. and Bobby J. Calder (1986, forthcoming), "Psychological Effects of Restaurant Meetings on Industrial Buyers," Journal of Marketing.

Taylor, S.E., and J. Crocker (1980), "Schematic Bases of Social Information Processing," in Social Cognition: The Ontario Symposium, eds. E.T. Higgins, C.P. Herman, and M.P. Zanna, Hillsdale, NJ: Erlbaum.

Vidmar, N. (1971), "Effects of Representational Roles and Mediators on Negotiation Effectiveness," Journal of Personality and Social Psychology, 17, 48-58.

Webster, Frederick E., Jr., and Yoram Wind (1972), Organizational Buying Behavior, Englewood Cliffs, New Jersey: Prentice-Hall.

Weick, Karl E. (1979), "Cognitive Processes in Organizations," in Research in Organizational Behavior, Vol. 1, ed. B.M. Staw, Greenwich, CT: Jai Press, 41-74.

----------------------------------------