Effects of Commercial Complexity, the Candidate, and Issue Vs. Image Strategies in Political Ads

Citation:

Joan Schleuder (1990) ,"Effects of Commercial Complexity, the Candidate, and Issue Vs. Image Strategies in Political Ads", in NA - Advances in Consumer Research Volume 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT : Association for Consumer Research, Pages: 159-168.

Advances in Consumer Research Volume 17, 1990      Pages 159-168

EFFECTS OF COMMERCIAL COMPLEXITY, THE CANDIDATE, AND ISSUE VS. IMAGE STRATEGIES IN POLITICAL ADS

Joan Schleuder, University of Texas

[This research was funded by the National Political Advertising Research Project, funded by the Gannett Foundation. Thanks go to Sandra Gaiser and the fall 1988 University of Texas Journalism Research Methods class for their help in preparing stimuli, running subjects and preparing data for analysis.]

In the 1940s, when the classic political communication studies were conducted, presidential campaigns would have been wise to severely limit their advertising expenses. The mass media were found to have almost no impact on attitudes, voting preferences and political knowledge. More than 80 percent of the voters decided who they would vote for before the campaign began and political party affiliation and interpersonal communication had the greatest effect on voting behavior (Lazarsfeld, Berelson & Gaudet, 1944; Berelson, Lazarsfeld & McPhee, 1954). But the "media" described by the classic studies did not include television.

Recent research indicates that many of today's voters identify themselves as people who vote for the candidate rather than the party and that the television plays a larger role in their voting decisions than interpersonal communication (Mendelsohn & Crespi, 1970; DeVries & Tarrance, 1972). Today campaign officials consider it wise to advertise, and they have been allocating ever increasing percentages of their funds to television advertising (Shyles, 1986). There's no doubt that campaigners consider television an effective means for conveying political information. Current research tends to back up this perception. Today's voters report learning about presidential candidates from television commercials, and they even say they "became better acquainted" with candidates from television advertising (O'Keefe & Sheinkopf, 1974; Mendelsohn & O'Keefe, 1976). Much research has found a high correlation between exposure to political ads and presidential candidate knowledge (Patterson & McClure, 1976; Atkin & Heald, 1976; Hofstetter & Buss, 1980). The purpose of this study is to assess whether television advertising presents political information in a way that allows viewers to learn more about candidates and store information that will be useful at the time a final voting decision is made. This work examines how different characteristics of actual 1988 presidential campaign commercials affected the way people thought about the candidates.

The first research question addresses whether commercials that emphasize issues elicit more attention and are better remembered than image-oriented ads. Next, this study asks whether the type of candidate advertised influences the effectiveness of the advertising strategy adopted and whether the candidate type and strategy interact to affect cognitive processing. Finally, the effect of overall commercial complexity and it's interaction with strategy and candidate type is assessed.

Issue vs. Image Strategy

Much political advertising research has focused on the clash between commercials that adopt the rational, issue-oriented, approach and those that adopt the image model (Hellweg, Dionisopoulos & Kugler, 1989). The results of some of these studies suggest that issue concerns are primary to voters (Rusk & Weisberg, 19726. But most studies have found image-oriented political messages more effective on measures such as perceptions of the candidate (Kraus & Smith, 1977; Dennis, Chaffee & Choe, 1979) and candidate preference (Marshall, 1983). Clark (1979) argues that to gain as many votes as possible, candidates must appear to address issues while running their campaign with an image-oriented strategy.

Natchez and Bupp (1986) found candidate image to be the best predictor of voter behavior and attributed this to the ease with which image information is processed. The authors concluded that thinking about issues is more difficult than thinking about images because issues need more information to support them. Image-oriented advertisements have also produced higher content recall scores than issue ads in an experimental setting (Kaid & Sanders, 1978). In addition, when survey respondents were asked to remember information presented in political advertisements, they most often reported image information (O'Keefe & Sheinkopf, 1974).

Few issue vs. image studies of political advertising directly address the effect strategy has on the cognitive processing of candidate information. Still, based on this review of the literature, it seems reasonable to predict overall higher attention and memory scores for image-oriented ads.

Candidate Type

The 1988 Presidential Campaign offered the unique opportunity to compare commercials for two types of candidates: A familiar national figure and a relatively unknown, novel politician. George Bush had been the Vice President of the United States for eight years prior to his election. He had spent many years in high visibility positions with the federal government. In addition, he emerged early in the campaign as the Republican Party front runner and received much more media coverage as a result than Michael Dukakis, the Democratic nominee. Dukakis may have been well-known in Massachusetts where he served as governor, but to the rest of the nation he was a political new-comer.

Generally, when a new product is introduced, a rational, information-oriented advertising campaign is used. The public needs to know what this new product has to offer. If a new political candidate is analogous to a new product, perhaps commercials advertising him (or her) should focus more on issues than on image. But for a well-known politician who has been around a long time, it's logical to assume that an image approach would work better, just as it works better for old, well established products like Coca-Cola.

People have more to learn about a novel candidate in a shorter period of time. Consequently, they may be motivated to attend to and remember more pieces of information about Dukakis than about Bush. If this is the case, highly complex Dukakis commercials -- especially if issue-oriented -should result in better cognitive processing scores.

Commercial Complexity

Natchez & Bupp (1986) thought that image ads were less complex than issue ads and that this lack of complexity lead to better processing of image commercial information. If Natchez & Bupp (1986) are correct, overall low complexity ads should result in better cognitive processing even if Dukakis issue-oriented ads specifically benefit from high complexity executions. Bush image-oriented ads that are low in complexity should be processed more effectively than any other type of ad.

In this study, complexity is a formal characteristic of commercials and is defined by the amount of visual and verbal information present in commercials per time unit. In a program of work that has investigated complexity's effects on attention and memory, Schleuder, Thorson and Reeves found that simple product commercials have elicited more attention than complex ones (Thorson, Reeves & Schleuder, 1985; 1987; Schleuder, Thorson & Reeves, 1987; 1988; Schleuder & Gaiser, 1989). Although this finding has been rigorous with product commercials, it would be interesting to see if political ads also elicit more attention and result in better memory when they are kept simple. Beginning to examine how a commercial formal feature, such as complexity, and content features, such as the advertising strategy adopted and the candidate promoted, is important to developing an understanding of how viewers process the information contained in television commercials.

Based on past research and speculation in the issue vs. image political advertising strategy literature, overall low complexity is expected to result in higher cognitive processing scores and offer special facilitation to issue-oriented advertisements for Bush. Dukakis advertisements about issues are expected to benefit from high complexity approaches.

Attention and Memory

Attention is an on-going process that refers to how viewers allocate mental resources to information in television messages (Schleuder, Thorson & Reeves, 1988). Mental resources refer to visual and auditory sensory systems and the memory mechanisms that determine how new information will be integrated with the old. This definition derives from Kahneman's (1973) general capacity model of attention and from multiple resources theory (Wickens, 1984) and rests on the assumption that the ability to attend to more than one simultaneous event is limited (Broadbent, 1958).

The secondary task reaction time method allows advertising and mass communication researchers to measure attention as an on-going process. In the typical secondary task experiment, viewers are told to concentrate on watching television, but to also respond as quickly as they can to periodic tones by pressing a reaction time button as quickly as possible. The assumption underlying use of the task is that performance on the secondary task decreases as attention to the primary task increases. As more attention is allocated to television viewing, less attention is left over to use in responding to the secondary task. To insure that this assumption holds, secondary task experiments should incorporate a manipulation check to assess speed-accuracy trade-offs. If subjects focus attention on the button-pressing task rather than television viewing, the reaction time measures become meaningless.

Visual and verbal recognition memory refers to the process by which information picked up from viewing commercials is stored in long-term memory and retrieved in response to a cue. Recognition is used, rather than recall because recognition methods focus on identifying the strength of the encoding process (Klatzky, 1980). The distinction between visual and verbal information retrieval is made because television is both a visual and verbal medium. In addition, as Garramone (1989) has suggested, auditory and visual processing schemes may differ in response to issue vs. image advertising strategies.

Hypotheses

The issue vs. image strategy adopted by the advertiser is a fundamental decision and, based on recent studies (O'Keefe & Sheinkopf, 1974; Kaid & Sanders, 1978; Natchez and Bupp, 1986), is expected to have an effect on attention and memory no matter how other political ad characteristics vary. Candidate type advertised also is expected to have an effect on how political information is processed. In addition, because novelty should lead to enhanced processing of ads for Dukakis (Reeves, Thorson & Schleuder, 1987), Candidate type is expected to interact with strategy and also with complexity. Commercial complexity is expected to affect how the information in political ads is processed. In addition, a low level of complexity is expected to enhance attention and memory for Bush issue strategy commercials while a high level of complexity is expected to enhance attention and memory for Dukakis ads about issues.

H1: Overall, attention as indicated by secondary task reaction time, and visual and verbal memory recognition scores are expected to be higher for Image Strategy commercials than for Issue Strategy commercials.

H2: Overall, attention, and visual and verbal memory are expected to be higher for Dukakis commercials than for Bush commercials.

H3: Overall, Low Complexity executions of ads are expected to result in higher attention, and visual and verbal memory than High Complexity executions.

H4: High Complexity Dukakis ads are expected to produce higher cognitive processing scores as indicated by attention, and visual and verbal memory scores, than Low Complexity Dukakis ads.

H5: Low Complexity Bush ads are expected to produce higher cognitive processing scores than High Complexity Bush ads.

H6: High Complexity Dukakis ads, especially those about Issues, are expected to produce higher cognitive processing scores than other types of Dukakis ads.

H7: Low Complexity Bush ads, especially those that emphasize the candidate's Image, are expected to produce higher cognitive processing scores than other types of Bush ads.

METHOD

SubJects

A total of 44 people participated. Twenty-one adults were recruited from churches and social groups in Travis County, Texas and their organizations were paid $5 for their participation. The 23 University of Texas undergraduate communication students participated for course credit. An additional 44 subjects (half were students and half were recruited from the community) participated as members of the secondary task manipulation check control group.

Stimuli

Eight 30-second political advertisements, four for George Rush and four for Michael Dukakis, were selected from 34 positive appeal commercials that had aired for the two candidates by October 1, 1988 (the time the experiment was run). The 34 ads were collected from the Democratic and Republican parties, from the Political Advertising Archive at The University of Oklahoma and by recording them off-air. Two commercials for Bush and two for Dukakis used issue strategies, and two for each candidate used image strategies. The Bush issue strategy ads included an ad suggesting Bush would "keep America at work," and a talking head execution in which Bush talked about the need for American trade to become more competitive with foreign countries. The Bush image strategy ads included an ad describing Bush's government experience and a biographical ad which stressed Bush's role as a family man. The Dukakis issue strategy ads included a union member talking about the need to create more jobs and a talking head form ad in which Dukakis discussed balancing the federal budget. The Dukakis ads that stressed image included Tipp O'Neill describing Dukakis' character and a biography of Dukakis' life. Eight advertisements were used in the experiment to enhance generalizability of the effects to other political advertisements (Jackson & Jacobs, 1983).

The Bush and Dukakis ads, along with eight product commercials, were embedded in a half-hour episode of the situation comedy Cheers. In order to keep the viewing experience as natural as possible, each subject saw only four of the eight political ads during their half-hour viewing session. One political ad was placed in each of four 90-second commercial pods. The order in which the political commercial fell in the pod (first, second or third) and the pod order (four orders) were counterbalanced to control for practice and fatigue effects. Which four ads each group of subjects saw was determined by their random assignment to one of two strategy conditions. Clearly audible secondary task tones were triggered by four cues that were placed at random intervals in each political and product commercial. Cues were also placed in the programming so that subjects did not suspect that the experiment focused on the advertisements.

Strategy and Candidate. The 22 subjects assigned to the Issue Strategy Condition watched the two Bush and two Dukakis ads judged to be the most prototypical examples of commercials about "economic issues important to the 1988 Presidential Campaign". There was no variation in the four coders' rating scores for these commercials. The ads used were ranked first (most prototypical) or second out of 17 Bush ads, and first or second out of 17 Dukakis commercials by all four coders. The coders were graduate students in an introductory research methods course.

The ads selected for the Image Strategy Condition were those judged, by the same coders, to be the most representative of ads for Bush and Dukakis that "focus on the image of the candidate as a leader." There was little variation in the rankings for the four advertisements selected. Three of the coders rankings were identical and the fourth coder included the same ads in the top rankings for each candidate, but the Dukakis ad rankings were in reverse order.

Complexity. Prior to being coded for advertising strategy, each of the 34 political advertisements was rated for complexity following the rating procedure used in several recent attention and memory studies (see Schleuder, Thorson & Reeves, 1988). Verbal complexity was indexed by coding the presence of music, voice-overs and natural sound and by counting the number of macro-propositions, or idea units). Messages with more types of sound and more propositions were considered more complex. Video complexity was assessed separately but combined with the verbal complexity score to determine each ad's overall complexity ranking. High video complexity scores were given to messages with many scene changes, dissolves and edits, much camera movement such as pans or zooms, much person and object movement and many superimposed images.

In addition, a group of 32 people provided subjective ratings of complexity during a pretest. These subjects were asked to view the 34 commercials and estimate on a 100-point scale the magnitude of complexity of the video, verbal or both channels. The subjective ratings provided a manipulation check on the complexity categories assigned to the political commercials and ensured that an ad rated low in complexity was perceived to be low in complexity to most viewers. Four of the eight ads used in this experiment were categorized as High Complexity and four were categorized as Low Complexity. The Bush Issue Strategy ad about keeping America working was a High Complexity ad, and his ad about foreign trade competition was Low Complexity. The Dukakis Issue Strategy ad about balancing the federal budget was High Complexity and his ad about the creation of new jobs was Low Complexity. The Bush Image Strategy ad about his life and role as a family man was High Complexity and the ad about his government experience was Low Complexity. The Dukakis Image Strategy ad about his life was High Complexity and Tip O'Neill's character appraisal was Low Complexity.

Experimental Procedure

Subjects were run one at a time after being randomly assigned to view either four Issue Strategy ads or four Image Strategy ads. Subjects were told that the purpose of the study was to see how much they could remember from watching television. They were also told to press the reaction-time button in front of them as quickly as they could whenever they heard a tone. Each subject "practiced" hitting the reaction time button in response to a tone before the experiment began so that 36 baseline simple reaction times could be collected and later used to assess individuals' basic motor speed. Then each subject viewed a brief episode of "Cheers" and a product commercial on a practice tape. They got practice responding to the tone secondary task and were given visual and verbal recognition tests about information in the practice program and in the commercial.

Subjects next viewed the one-half hour situation comedy and commercials. They were given visual and verbal recognition tests following the viewing period. The tests asked questions about "Cheers" and about product and political advertisements. Subjects then answered several questions about the upcoming presidential election including their intention to vote and their feelings about the two presidential candidates. The procedure for the 22 subjects in the secondary task manipulation check control group was identical except there was no secondary task.

Apparatus

The subjects watched Cheers, the embedded commercials and video-frame recognition tests on a 19-inch Panasonic 2010 video monitor placed five feet from their chair. The advertisements and programming were played on a Panasonic 6500 videotape recorder. An IBM AT microcomputer, with Labpac and Timecode reader/generator cards, controlled the tone presentation and the presentation of stimuli. Reaction-time responses were collected by a button box built by university technical staff.

Design

The study was a 2 X 2 X 2 mixed-factors design. Advertising Strategy was a between-subjects factor with two conditions: Issue vs. Image Strategy. Candidate was a within-subjects factor with two conditions: Bush vs. Dukakis. Complexity was a within-subjects factor with two levels Low vs. High. For the attention-measure, there were four repetitions of reaction-time nested in each cell of the Advertising Strategy X Candidate X Complexity factorial design.

There were also two between-subjects control factors, each with three levels -- commercial pod order and political commercial position within each pod order. There was also a secondary task vs. nosecondary task between-subjects factor included in the design as a manipulation check so that speed-accuracy trade-offs on the reaction time and memory test tasks could be detected if they existed.

Dependent Measures

Attention. The secondary task reaction time method was used as an indicator of how intensely viewers paid attention to the political commercials. Subjects were instructed to pay primary attention to the television, but also to press a reaction time button as quickly as they could in response to periodic tones. Slower reaction time on the button pressing task indicated that subjects paid more attention to viewing. When the intensity of viewer attention to the political commercials was low, subjects had more attentional resources left over to use in hitting the reaction time button and reaction time scores were faster. Attention scores were expressed in msec. units.

Visual memory was measured for each political advertisement by frame recognition tests administered after the 30-minute program was complete. The political ad was identified and six 1/15-second frames were flashed on the screen consecutively. Some of the frames were images that subjects had seen in the political ads and some were not. The number correct out of six frame identification items was the score for each political commercial.

Verbal memory was measured by giving subjects a six-item multiple-choice test for each political commercial. This test was administered after the visual recognition test. Each question offered four alternative choices. The number correct on the six-item test was the verbal recognition score for each news story.

FIGURE 1

MEAN REACTION TIME AS A FUNCTION OF STRATEGY, CANDIDATE AND COMPLEXITY

Manipulation Check for Speed-Accuracy Trade-offs

When the pattern of memory scores for subjects in the secondary task condition were compared to those in the no-task control condition, speed-accuracy trade-offs were not apparent. If subjects were trading speed on the button-pressing task for accuracy on the memory test, the memory scores would be significantly better for the control subjects than for the secondary task subjects -- this was not the case for visual recognition tests [F(1.84) = .00, g = .96] or for verbal recognition scores [F(1.84) = 3.28, n < .08].

RESULTS

Attention

H1 predicted that Image Strategy ads would elicit more cognitive processing than Issue Strategy ads. This hypothesis was supported by the results of the 2 (Issue v Image) X 2 (Bush v Dukakis) X 2 (Low v High Complexity) mixed-factors Analysis of Covariance for attention scores (F(1,41) = 4-91, p < .05). Mean reaction time in msecs. was 329 for Issue ads and 392 for Images ads.

ANCOVA was used only when between-group reaction time comparisons were made. People vary tremendously in their basic motor skills. Controlling for this by collecting baseline reaction time prior to the beginning of experiments allows the variation due only to differences in individuals' motor skills to be removed before differences due to the experimental manipulation are assessed.

There was also a three-way interaction among Strategy, Candidate and Complexity factors (F(1,41) = 5.66,; p< .05), as shown in Figure 1. Attention paid to Bush commercials was not affected by Complexity in either Strategy condition. However, Duke commercials' ability to elicit viewer attention was facilitated by Low Complexity when ads stressed the candidate's Image, but by High Complexity for ads that stressed Dukakis' stands on economic issues, as predicted by H6. Support for H2, H3, H4, H5, and H7 was not produced by the attention data.

Visual Memory

H1, which predicted higher cognitive processing scores for Image Strategies, was not directly supported by the visual recognition findings. The 2 (Issue v Image) X 2 (Bush v Dukakis) X 2 (Low v High Complexity) mixed factors Analysis of Variance did not produce a main effect for Strategy.

As predicted by H2, overall visual memory was better for Dukakis than for Bush commercials (F(1.42) = 9.40, p < .005), with a mean score of 3.81 out of 6 for Dukakis and 3.34 for Bush. The Candidate and Complexity factors interacted (F(1,42) = 4.92, p < .05) in the manner predicted by H4 and H5 (see Figure 2).

Strategy interacted with Candidate and Complexity and there was a three-way interaction. As shown in Figure 3, Issue and Image Strategies work equally well for Bush ads. But, as predicted by H6, the visual elements of the Dukakis Issue ads were remembered better than the Image visuals (F(1.42) = 10.34, p < .005). Figure 4 shows the Strategy X Complexity interaction. Counter to H4 and H5 predictions, visual recognition scores for Image Strategy ads were not affected by Complexity level, and scores for Low Complexity Issue ads were slightly higher than for High Complexity Issue ads.

FIGURE 2

VISUAL RECOGNITION NUMBER CORRECT AS A FUNCTION OF CANDIDATE AND COMPLEXITY

FIGURE 3

VISUAL RECOGNITION NUMBER CORRECT AS A FUNCTION OF STRATEGY AND CANDIDATE

FIGURE 4

VISUAL RECOGNITION NUMBER CORRECT AS A FUNCTION OF STRATEGY AND COMPLEXITY

The three-way interaction (F(1,42) = 40.26, p < .001), shown in Figure 5, suggests that, counter to predictions, visual memory for Dukakis ads did not vary as a function of Strategy or Complexity. In addition, High Complexity Bush ads about Issues resulted in much lower visual memory scores than any other Bush ad approach.

Verbal Memory

The 2 (Issue v Image) X 2 (Bush v Dukakis) X (Low v High Complexity) mixed-factors ANOVA produced the predicted main effect for Strategy. Verbal memory for Image Strategy ads was better than for Issue Strategy (F(1,42) = 5.89,;p < .05), with mean scores of 2.81 out of 6 for Issue and 3.43 for Image ads. Strategy did not interact with any other experimental factor, but, as predicted by H4 and H5, the Candidate and Complexity factors interacted (F(1,42) = 14.76, p < .001). As shown in Figure 6, verbal memory scores were higher for High Complexity Dukakis ads than for Low Complexity Dukakis ads, and scores were higher for Low Complexity Bush ads than for High Complexity Bush ads.

DISCUSSION

The primary purpose of this study was to see if two important content characteristics of political ads -- the issue vs. image strategy and the candidate type -- interacted with commercial complexity to affect information processing. In general, the moment-to-moment attention results and the visual and verbal- memory scares supported the idea that strategy, candidate and complexity are important ad characteristics that can be manipulated in a manner that will result in optimal cognitive processing of political information.

Image-oriented ads elicited more attention than issue ads. Their verbal message was also better remembered. Visual memory was better for ads about Dukakis, especially Dukakis issue ads. In general, both visual and verbal memory were better for high complexity Dukakis ads and for low complexity Bush ads.

Viewers paid more attention to high complexity issue ads for Dukakis as expected. This is probably because a novel political candidate, like a new product, benefits from a rational promotion approach and a high level of motivation on the part of the viewer to learn what the new entity has to offer. Low complexity image ads for Bush also elicited a lot of attention, probably because Bush is an old, familiar national politician. Like many familiar products, he benefitted from ads that pitched his image with simple executions. This would give viewers plenty of time to devote to thinking about old memories they have about Bush (see Thorson, et al., 1985), and it's possible that this active processing of simple, familiar information is what causes the high level of attentional intensity.

Visual memory for political commercial information is highly specialized. Viewers remembered the visual information about Dukakis issue ads and Bush image ads best. Low complexity issue ads worked best. Visual memory was best for low complexity Bush ads about issues and for high complexity Bush ads about images. It appears that there is a mental trade-off process, or a time-sharing function, at work between the visual sensory system and attentional mechanisms when interactions among all three ad characteristics are examined. This type of trade-off has been observed between auditory and visual sensory systems, between auditory and attentional systems and between visual and attentional systems (Thorson, et al, 1985; 1986; Schleuder, et al., 1987; 1988). Identifying exactly what commercial characteristics trigger the cognitive time-sharing observed in this study should be a goal of future research.

FIGURE 5

VISUAL RECOGNITION NUMBER CORRECT AS A FUNCTION OF STRATEGY, CANDIDATE AND COMPLEXITY

FIGURE 6

VERBAL RECOGNITION NUMBER CORRECT AS A FUNCTION OF CANDIDATE AND COMPLEXITY

REFERENCES

Atkin, Charles & Gary Heald (1976), "Effects of political advertising," Public Opinion Quarterly, (Summer).

Berelson, Bernard, Paul Lazarsfeld & William McPhee (1954), Voting: A study of opinion formation in a presidential election, Chicago: University of Chicago Press.

Broadbent, Donald (1958), Perception and communication. London: Pergamon Press.

Clark, T.D. (1979), "An exploration of generic aspects of contemporary American campaign orations." Central States Speech Journal, 30, pp. 122-33.

Dennis. J.. Steven Chaffee & Sun Choe (1979), "Impact on partisan, image and issue voting, S. Kraus (Ed.), The great debates: Carter vs. Ford. 1976. Bloomington, IN: Indiana University Press.

DeVries, Walter & Lance Tarrance (1972), The ticket-splitter: A new force in American politics. Grand Rapids, Michigan: William B. Eerdmans.

Hellweg, Susan & George Dionisopoulos & Drew Kugler (1989), "Political candidate image: A state-of-the-art review," in B. Dervin & M.J. Voigt (Eds.) Progress in communication sciences Vol. 9. Norwood, New Jersey: Ablex Publishings Corporation.

Hofstetter, Richard & Terry Buss (1965), "Politics and last minute political television," Western Political Quarterly, 33, (March), pp. 24-37.

Garramone, Gina, Michael Steele, & Bruce Pinkleton (1989), "The role of cognitive schemata in determining candidate characteristic effects," paper presented at the Association of Consumer Research Annual meeting, New Orleans.

Jackson, Sally & Scott Jacobs (1983), "Generalizing about messages: Suggestions for design and analysis of experiments." Human Communication Research, 9, 169- 181

Kahneman, Daniel. (1973), Attention and effort. New York: Holt.

Kaid, Linda & Keith Sanders (1978), "Political television commercials: An experimental study of type and length." Communication Research, 5, 57 -70.

Klatzky, Roberta (1980), Human Memory: Structures and Processes. San Francisco: W.H. Freeman and Company.

Kraus, Sidney & R.G. Smith (1977), "Issues and images," in S. Kraus (Ed.), The great debates: Kennedy vs. Nixon, 1960. Bloomington, IN: Indiana Press, pp. 289-312.

Lazarsfeld, Paul, Bernard Berelson & Hazel Gaudet (1944), The people's choice. New York: Columbia University Press.

Marshall, T.R. (1983), "Evaluating presidential nominees: Opinion polls, issues, and personalities." Western Political Quarterly, 3, pp. 61-83.

Mendelson, Harold & Irving Crespi (1970), Polls, television and the new politics. Scranton: Chandler Publishing Co.

Mendelson, Harold & Garrett O'Keefe (1976), The people choose a president: Influences on voter decision Malung, New York: Praeger.

Natchez, P.B. & I.C. Bupp (1968), "Candidates, issues, and voters." Public Policy, 17, pp. 409437.

O'Keefe, Timothy & Kenneth Sheinkopf (1974), "The voter decides: Candidate Image or Campaign Issue?" Journal of Broadcasting, 18, pp. 403-12.

Patterson, Thomas., and Robert McClure (1976) The Unseeing Eye: The Myth of Television Power in National Elections, New York: G.P. Putnam.

Reeves, Byron, Esther Thorson & Joan Schleuder (1986), "Attention to television: Psychological theories and chronometric measures." In J. Bryant & D. Zillmann (Eds.) Perspectives on media effects (pp. 251-280). Hillsdale, New Jersey: Erlbaum.

Rusk, J.G. & H.F. Weisberg (1972), "Perceptions of presidential candidates: Implications for electoral change." Midwest Journal of Political Science, 16, pp. 388-410.

Schleuder, Joan (1989), "The effect of political message form on cognitive processing: TV commercials vs. news stories," an unpublished University of Texas, College of Communication Cognitive Research Laboratory manuscript.

Schleuder, Joan & Sandra Gaiser (1989), "How complexity interacts with content features of televised political messages to affect cognitive processing." An unpublished University of Texas, College of Communication Cognitive Research Laboratory manuscript.

Schleuder, Joan, Glen Cameron & Esther Thorson (1989), "Priming effects of television news teasers on attention to and memory for emotion-eliciting commercials." 1989 American Academy of Advertising Proceedings.

Schleuder, Joan, Esther Thorson & Byron Reeves (1988), "Effects of time compression and complexity on attention to television commercials," a top paper presented to the Mass Communication Division of the International Communication Association, New Orleans.

Schleuder, Joan, Esther Thorson & Byron Reeves (1987), "Effects of complexity and scene reordering on attention to television messages," paper presented to the Mass Communication Division of the International Communication Association, Montreal, Canada.

Shyles, L. (1986), "The televised political spot: Its structure, content, and role in the political system," In L. Kaid, D. Nimmo, & K. Sanders (Eds): New perspectives on political advertising. Carbondale, IL: Southern Illinois University Press.

Thorson, Esther, Byron Reeves & Joan Schleuder (1985), "Message complexity and attention to television." Communication Research, 12, 427-454.

Thorson, Esther., Byron Reeves & Joan Schleuder (1987), "Attention to local and global complexity in television messages." In M. McLaughlin (Ed.), Communication Yearbook 10. Beverly Hills: Sage.

Wickens, Christopher (1984), "Processing resources in attention." In R. Parasuraman & D. Davies (Eds.), Varieties of attention (pp. 63-102). New York: Academic Press.

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