There is a fundamental distinction between the kinds of emotion we all experience in our day-to-day lives and the kinds of emotion produced by a work of art. If advertising is an art, like music and poetry, painting and photography, novels and film, a discussion of the nature of effective advertising must deal with the experience of aesthetic emotion.

In life, thought and emotion usually come to us separately. From behind the curtain screening our unconscious mind our emotions dance and flicker like the light of a television seen from the street. In the blink of an eye, surprise turns to hatred; contentment turns to sadness; annoyance turns to affection. A chance look by an acquaintance triggers the chill wind of an emotion that we may reflect on days later, trying to understand what it means to us. Meanwhile, the mind is filled with fragmentary sound bites, as we channel surf our thoughts and images. Like Molly’s soliloquy from Joyce’s Ulysses, our interior monologues stream along by serendipitous processes of free association.

In art, thought and emotion are fused. The emotions generated in the audience by the work are there by design. The thoughts evoked in the audience by the work are there, too, by design. What is it exactly that a fine work of art does? It creates meaningful emotion. That is the definition of aesthetic emotion—it’s meaningful emotion, where feeling and thinking come together.

Robert McKee, a famous Hollywood scriptwriting consultant, writes about the “epiphanies” of storytelling:

“In short, a story well told gives you the very thing you cannot get from life: meaningful emotional experience. In life, experiences become meaningful with reflection in time. In art, they are meaningful now, at the instant they happen.”

McKee goes on to talk about the importance of the “controlling idea” for a film, which is “One clear, coherent sentence that expresses a story’s irreducible meaning.” In terms of advertising, it’s the idea that organizes the emotions generated by an ad.

Knowing the controlling idea for a film is critical for making editing decisions. What is the criterion by which the director decides that a precious scene is unnecessary, or too much? No matter how delightful or well written or superbly acted it is, if a scene does not advance the story, if it does not help express the controlling idea, it violates the artistic integrity of the work and should be left on the cutting room floor.

In advertising terms, the controlling idea for a creative execution is the strategy.

Much of the current discussion on the topic of emotional advertising versus rational advertising implies that an effective ad can be one type or the other. But an ad that creates an emotional charge without a thought attached, or that conveys a thought without an emotion attached, cannot be effective—for either without the other lacks meaning. Effective advertising builds brands by making products and services meaningful to the lives of consumers.

It is simply not enough, for instance, to try to determine advertising effectiveness by simply measuring how “enjoyable” an ad is. The enjoy-ability of an ad can be created cheaply through borrowed interest—attracting attention by telling a joke that has nothing to do with the brand—or it can be accomplished through the hard creative work of dramatizing an emotional idea that illuminates a core value of a brand.

Emotion that is organized by thought is like the idea of working energy in physics—it is the emotion generated by the ad that is on strategy. Aesthetic emotion, in a sense, can be thought of as the working emotion of an ad—the emotion that’s put to work to build a brand.

Aesthetic emotion, as we will show with research, is what creates the long-term effects of advertising.

An Experiment in Long Term Ad Effects

To better understand how emotion works in television advertising, we recently asked Unilever to participate in one of our R&D efforts. Our purpose was to study in detail how a set of TV commercials had worked over a long period of time to build a successful image for one of Unilever’s brands. With their cooperation, we conducted the following simple experiment.

In selecting a brand for study we had two criteria. First, we wanted a heavily advertised brand that was clearly successful in the marketplace—so that we could reasonably infer from the brand’s success that, at least on average, the advertising had indeed been effective. Second, we wanted an established brand, not a new product; one that had been around for a middling length of time so that it would be practical to re-test a major portion of the brand’s early advertising history.

The brand that we chose met these criteria was Degree deodorant. It was launched in the U.S. approximately a decade before this study was conducted; it is currently the number two brand in its category; and it is still gaining market share.

From the historical reel of advertising for Degree, we chose eight previously-aired television commercials for this research. These commercials represented a significant span of the early advertising history of this brand. Importantly, at the time of this test, all of these commercials had been off air for five years or more—so what we were studying were the residual long-term effects of this advertising.

For our experiment, we interviewed a total of 300 consumers. The design of the research was quite simple. Respondents were divided into two test cells of 150 respondents each (which were matched in terms of demographics and brand usage).

The first cell replicated a campaign pre-test, where respondents were shown all eight commercials together and then questioned, using picture sorts, about the parts of the commercials that made them “think” as well as the parts that made them “feel.”

The second cell replicated a campaign post-test or tracking study. Respondents were not shown any advertising at all. Instead, they were shown only the pictures from these old ads to probe the residual imagery that remained linked to the brand in their long-term memories.

By comparing results from the two cells using the common picture sorting methodology we were able to measure directly whether the thinking parts or the feeling parts of the commercials had contributed more over time to building the image of the brand.

Measuring the Campaign Imagery in the Pre-test

To represent the visual content of these eight ads we drew a sample of 98 image frames from the advertisings. Because we wanted to use the same pictures in the second de-branded tracking cell, this set of pictures excluded images that identified brand, such as package shots or visuals containing the brand name, logo, tag line or other explicit brand identifiers.

After respondents first performed our standard Flow of Attention® sort, where they were asked to sort the images based on those they remembered and those they didn’t, they were asked to sort pictures they remembered on two additional dimensions of response to the advertising. First, they sorted based on how much each image from the eight commercials made them think as they watched the commercial. The pictures were sorted into five ratings categories ranging from “made me think a lot” to “did not make me think at all.” Next, respondents sorted based how much each image made them feel as they watched the commercial, from “made me feel a lot” to “did not make me feel at all.”

When we began our analysis of these data, we quickly discovered that the ratings of thinking and feeling were highly correlated—which is perhaps not surprising in self- report data when you ask consumers to try to unravel their thoughts and emotions as they attempt to reconstruct their experience of an ad. Therefore, to create independent measures for proper statistical analysis, we created two new variables.

First, we created a new variable by taking the sum of the two ratings, think + feel. This variable measures the total response to images where consumers perceived both a thought and an emotion in their reactions to the ad. In a sense, this variable measures the total intensity of consumer response.

We created the second variable by taking the difference of the two ratings, think – feel. This variable measures consumers’ perceived imbalance in their response to an image, either favoring rational thought over emotional content or the converse.

These two new variables are indeed independent (r-squared = .06) and so, for the purposes of our analysis, provided us with meaningful consumer-based measures of both the union and the separation of the thinking and the feeling that the consumer experiences when watching commercials.

Measuring Brand Imagery in the Post Test

In the second cell, following the design of a “tracking” study, respondents were not re-exposed to any of the test commercials. The purpose here was to identify visual images from the old advertising which, five years after the last ad had aired, were still linked in the minds of consumers to the brand.

So, in this part of the study, the same 98 pictures which were used in pre-test cell–as well as an additional 20 control images from other advertising– were used as a visual vocabulary for probing the current image of the brand. The randomized deck of pictures was sorted by consumers according to various criteria of brand association.

For example, to measure brand image, consumers were asked to rate how well each picture fit with their perceptions of the kind of person or situation that they associate with the brand. Or, to measure the fit with the brand’s positioning in the marketplace, respondents were asked to rate how well each visual fit with the strategic idea expressed in a sentence describing the brand’s long-term unique selling proposition.

Thought and Emotion Come Together in a Brand

For the next step in our experiment we compared the picture ratings from the campaign pre-test cell to the pictures associated with the brand from the tracker cell. What did we find out?

The images from the advertising campaign that made consumers simultaneously both think and feel were much more likely to be associated with the brand’s current image. These same images also better fit the brand’s positioning in the marketplace.


Images that engendered a thought without an emotion—or an emotion without a thought—were not associated with the brand to any significant degree. You can see this in the correlations between the two sets of data shown in Exhibit 1.

The concepts of brand image and a brand’s positioning are the yin and the yang of classical brand theories. Brand image, first popularized in the fifties by the ad man David Ogilvy, describes a brand in terms of associations—the set of images, emotions and experiences linked in the mind of the consumer to a given brand name. A brand’s positioning, popularized by the brand consultants Al Reis and Jack Trout in the seventies, defines a brand in terms of the verbal concepts that differentiate it within a relevant frame of reference of competitors.

In the terms of modern psychology, a brand’s image is stored in the episodic memory system—for that is where the mind stores the sensory images and emotions of personalized experience. In contrast, a brand’s positioning is stored in the semantic memory system—for that is the part of the mind that deals with semantic propositions and rational concepts.

At first blush it appears that these two views of brands could not be further apart. One is an emotional view of the brand; the other is a rational view of the brand. One has an inner-directed focus; the other has an outer-directed focus. Both views are undoubtedly correct.

This research finding shows that the working parts of an ad are those that express the dual nature of brands. When advertising forges a link between these two systems of the mind, thought and feeling unite to form the meaning of the brand.

The Flow of Audience Attention

The importance of rhythmic structure is well known to cognitive psychologists. In conversation, for example, rhythm and beat in language help us organize the information in a face-to-face exchange to avoid cognitive overload by focusing on predictable, strategic moments to pay attention. Importantly, researchers have found that the beat of a conversation is when speakers often convey key information and bring up new topics.

For actors, a beat is an exchange of an action/reaction in character behavior. A good actor might be described as “not missing a beat” in improvising a reaction to an unexpected event in a play.

Robert McKee defines the “beat” as the smallest unit of film structure. In terms of film storytelling, however, a beat is the exchange between a director and the audience as together they co-create the meaning of the film.

In pre-testing television commercials we measure the moment-by-moment flow of audience attention through the film. The rhythmic appearance of peak moments in the flow has been shown to be highly predictive of commercial effectiveness both in terms of overall breakthrough and branding. Describing the best remembered moments from the film in terms of visual “peaks” is simply different language for talking about the beat of the film.

For those of you who are unfamiliar with this Ameritest technique, our Picture Sorts™ diagnostics have been extensively validated against the report card measures of all of the major copytesting systems used by advertisers in the U.S. For example, the Flow of Attention® is one of the most powerful diagnostics for the Millward Brown performance measures comprising the Effectiveness Index.

In a rough sense, the Flow of Attention (FOA) might be thought of as a much more nuanced and powerful version of the traditional verbatim report of the creative elements that “stand-out” in the story of an ad. The power of the FOA is that it’s an aided rather than an unaided measure of stand out that captures visual information from every respondent about every discrete visual element of an ad. This is particularly important with creative imagery that respondents have trouble putting into words.

Peak moments are the outcroppings of the experience the audience has of the commercial. What our experiment shows is that these are the high points in the ad for both thinking and feeling, as you can see in Exhibit 2. Audience attention is focused on the most meaningful moments in the ad—the moments where thought and emotion come together.

An important function of memory is that it enables us to forget. Given our limited capacity to keep important information in our conscious mind we need a capacity to forget. The valleys in the Flow of Attention are literally the images that the audience has begun to forget within twenty minutes of seeing the commercial.

That is not to say that these other “forgotten” moments in the commercial are unimportant. They may be necessary for the narrative build in the emotions and thoughts created by the moving images, or a set up for a situation. Valleys in the Flow of Attention are the scaffolding for the experience that the audience has of the advertising, which is then removed from memory after the tower of meaning is construct. To see this, we can look at the results from a question in our tracking experiment where we asked consumers which images they recognized ever having seen in Degree’s television advertising. From across the test eight commercials, we took images from the peak moments of audience attention, from the valley’s of attention, and, as a control, images that were never in Degree’s advertising but which were taken from other personal care product commercials.

As you can see, images that lie on the peak moments of the Flow of Attention are remembered at significantly higher levels than the “false awareness” levels represented by the control images. Non-peak images do no better than the control images. In other words, only the images from the peaks are stored in long-term memory.

In many marketing applications, the Pareto principle applies—for example, the idea that twenty percent of your products account for eighty percent of your profits. We see here that the 80/20 rule applies to advertising film as well—the peak moments in the film do the long term work of building brand image.

One interpretation this finding suggests is that Picture Sort peaks identify those moments in an ad where an intense emotional experience has been created—one that the mind of the consumer simultaneously stores in both the episodic and the semantic parts of the brain. In scientific terms we might say that meaning is created by mental processes that “map” our emotions onto our thoughts. Aristotle once wrote that “the soul never thinks without an image.” To turn this thought on its head, this pattern of findings suggests that an image unattached to a thought cannot properly transform into the meaning of the brand.


There are a number of practical implications from this simple experiment for marketing practitioners. For instance, the finding just reported has serious consequences for tracking advertising awareness, particularly as more and more advertisers are moving to on-line recognition-based measures of ad awareness. We know from other in-house research that even experienced researchers have a difficult time predicting what will be the Flow of Attention peaks for the consumer. Yet if the wrong imagery is chosen from the advertising to create the awareness stimulus—using images from the valleys instead of the peak moments of the ad—on-line trackers might seriously understate the true in-market awareness of the advertising, and hence the advertising ROI.

More generally, however, this experiment questions the underlying assumptions that fuel the debate about emotional advertising versus rational advertising. We would argue that this debate is a red herring. All effective television commercials contain focused moments where thought and emotion come together. In literary terms, from the standpoint of how advertising works to build a brand, the consumer remembers the poetry and discards the prose of the commercial experience.


Young, Charles E. and Michael Robinson, “Video Rythms and Recall”, Journal of Advertising Research, March/April, 1992.

Young, Charles E. and John Kastenholz, “Emotion in TV ads”, Admap, January, 2004, Issue 442, pp 40–42.

Young Charles E., “Capturing the Flow of Emotion in Television Commercials: A New Approach”, Journal of Advertising Research, June 2004, Volume 44, NO. 2, pp 203–209.

McKee, Robert, Story: Substance, Structure, Style and The Principles of Screenwriting, published by Regan Book/HarperCollinss; 1st edition (December 17, 1997)