Upload
ngohanh
View
225
Download
0
Embed Size (px)
Citation preview
INTERMEDIATE ADVERTISING EFFECTS:
THE MAC MODEL
Tim Ambler
Centre for Marketing Working PaperNo. 96-906
Revised edition, May 1997
Tim Ambler is Grand Metropolitan Senior Fellow at London Business School. He is indebtedand grateful to Steven Rose, Tom Burne, Andy Ioannides and David B. Wolfe for theirconnections to neurobiology and to colleagues at London and Columbia Business Schools,especially Paddy Barwise, Sri Devi Deepak, Kent Grayson, Kathy Hammond, MorrisHolbrook, Flora Kokkinaki, Don Lehmann, Gil McWilliam, John O’Shaughnessy, TomRobertson, Arvind Sahay, Chris Styles, Demetrios Vakratsas and Chris Walters for theircontributions and perceptive comments on drafts.
London Business School, Regent's Park, London NW1 4SA, U.K.Tel: +44 (0)171 262-5050 Fax: +44 (0)171 724-1145
[email protected] http://www.lbs.ac.uk
Copyright © London Business School 1997.
Intermediate Advertising Effects:
Centre for Marketing
�
1
The MAC Model
Abstract
Advertising was once believed to operate wholly as information transference. For the last 100
years, advertising has also been seen as persuasive with cognition continuing to dominate
advertising research. The role of affect in advertising has a stream of research not fully
integrated with the Cartesian (cognitive) stream. Neuroscience may provide the linkage. This
paper suggests that consumer behavior is driven primarily by memory (M), which includes
habits and conditioning, and secondarily by feelings (affect - A). The utility maximising role
of reason (cognition - C) typically plays a minor, albeit important, role. The M ⊃ A ⊃ C, or
MAC, model of the intermediate effects of advertising (“⊃“, includes, is used in the sense of
“setting the context for”) may be part of a paradigm shift in marketing research toward more
emphasis on affect. Brand equity is presented as ΣM, i.e. the combined brand memory of the
relevant population. The paper concludes with managerial implications and proposals for
testing the MAC model.
1
Intermediate Advertising Effects:
The MAC Model
Advertising may change consumer behavior through the provision of information and/or
persuasion and/or by preventing the erosion of loyalty. Advertising may reassure consumers
whose decisions are already made and it may permit a brand to maintain a higher price. This
paper is concerned with the mind of the consumer, with the intermediate effects that take place
between the consumer’s perception of the advertising and any such behavioral changes1. We
know that, with the possible exception of direct response and point of sale advertising, these
intermediate effects must exist if only because of the interval of time between the advertising
and the next behavioral change opportunity, e.g. shopping.
Marketers have been seeking to model these intermediate effects since AIDA (Attention Ö
Interest Ö Desire Ö Action) first appeared 100 years ago (Strong 1925, p.76). Before that,
advertising was seen simply as providing information. The information might be true or false,
and often exaggerated (Turner 1952), but intermediate effects were not considered.
2
This paper sets out to reconcile what we know from the marketing literature about
intermediate effects with what we know from the neuroscience literature about mental
processes. Three factors are mostly concerned: cognition, affect and memory which will,
following Holbrook (1986) be abbreviated as C, A and M. “Experience” is related to memory
in the marketing literature but ambiguously as it may also refer to actual behavior (B) as distinct
from antecedent and consequential mental conditions. “Memory” is not new to the marketing
literature, e.g. Srull (1990), Alba et al. (1991), though it is generally used in the context of
brand and advertising recognition and recall, i.e. conscious memory. Memory will later be
defined to include all forms of mental storage such as facts, procedures, habits and other
nonconscious conditioning which result from experience in the sense of behavior.
From these streams of literature, a model of how advertising works is formed. The model may
perhaps be extended to all marketing decision-making to position cognition within the context
of memory and then affect. The perception of cognition as the dominant mental process may be
traced to “cogito ergo sum”, if not the ancient Greeks, though cognition (knowing) is not
strictly equivalent to thinking (cogitation). Since “cognition” is commonly (e.g. Holbrook et al.
1990) associated with all rational aspects of the mind, it is so used here. “Affect” includes
emotions and feelings, such as desire, instinct and volition (conation). Cognition and affect
are, for the purposes of this paper, interactive but mutually exclusive.
1 To save redundancy, “change” hereafter includes “maintain” if, without the advertising,
3
Cartesianism, i.e. utility maximizing rationality and the exclusive focus on cognitive processes,
has been “a basic assumption in marketing” (O’Shaughnessy 1992, p.90). The interest now
being taken in affective responses, relationship marketing, post-modernism and other non-
cognitive explanations of decision-making may be part of a paradigm shift to “Post-
Cartesianism”. Hunt (1992) is only one researcher who has drawn attention to the divisions
between the rationalists and the relativists. I explore the extent to which neuroscience, which
meets his requirements for objective science but also deals with affective and social issues
which concern the relativists, may link the two schools. Managerial implications and proposals
for testing the MAC model are provided, before drawing conclusions.
Marketing literature
This section briefly follows the development of consumer behavior research from Cartesian
assumptions to non-cognitive approaches. Research into the intermediate effects of advertising
is then summarised before concluding with the suggestion that cognitive bias permeates the
research process.
mental states or behavioral patterns would have eroded.
4
Consumer behavior
The Cartesian perspective has been termed “man as computer” (Holbrook 1986; Hirschman
1993) or the “information processing model” (Bettman 1979) and assumes consumers to be
rational utility maximizers (e.g. Howard and Sheth 1969; Engel, Blackwell and Kollat 1978)
who follow a series of logical steps (e.g. problem recognition, situation assessment, short-
listing of alternatives, establishment of choice criteria, evaluation and selection). More recently,
products have seen as bundles of attributes which the consumer instinctively assesses, e.g.
conjoint analysis (Green and Wind 1975; Green and Srinivasan 1990). Such models assume
that the consumer has a short list (evoked set - Howard 1977; Nedungadi 1990) of alternative
choices (products/services) each of which can be analyzed into discrete but comparable
attributes.
As it became clear that consumers do not fully engage cognition in trivial and repetitive
decisions (e.g. choice of brand of soup), the approach was modified to accommodate
routinization (Chaffee and McLeod 1973; Robertson 1976). Loyalty may reflect the
convenience inherent in repetitive behavior, rather than commitment to the brand purchased or
active decision making. Krugman (1965, 1967) introduced involvement to distinguish those
situations, where minds are active, from those where only the auto-pilot is engaged. Krugman
noted that learning can take place without involvement. Petty and Cacioppo (1981)
distinguished central (extensive cognitive processing) from peripheral (limited) processing of
advertising information.
5
Anderson (1983) proposed “a common cognitive system for higher level processing” (p.1), i.e.
a universal architecture which included both central and peripheral formats, and rejected the
notion of the mind having separate organs for separate functions (p.4). The mind, in his view,
was a unitary production centre though memory had three functions: short-term working,
factual (declarative) and rules (procedural) storage. Anderson’s ACT* computer-like model
has no role for affect.
The dominance of cognition has been implicitly and explicitly challenged from various
directions (e.g. Batra 1986; Holbrook 1986). Zajonc (1980) showed that “affective reactions
can take place without extensive perceptual and cognitive encoding”. Group decisions can be
more reliable than individual ones because of affective reactions (p.171). The interplay
between individual feelings, social behavior and decision-making is one of Zajonc’s key
findings; cognition had little, if any, role.
No consumer cognitive brand selection process may exist at all, even originally to add a brand
to the evoked set from which consumers pick (Kassarjian 1978; Olshavsky and Granbois
1979). People do not consider options; they think only when they have to (Kagan 1989;
Csikszentmihalyi 1993).
6
Pham (1996) showed that affect is a source of information in certain buying decisions, e.g.
buying a dress, because the customer anticipates the pleasure of the purchased item and this
prompts the shopping expedition. He distinguishes this process from a rational evaluation of
options: “affect recruitment” involves the immediate experience of the anticipated feelings.
Feminist writers (e.g. Hirschman 1993) challenge the dominant cognitive approach as
androcentric. The suggestion is that, if more women had been involved, consumer research
would have been more sensitive to affect, with less acceptance of the machine metaphor
(cognition). The rejection of logic is a dangerous road for an academic but there is, as we will
see, substance to Hirschman’s claim that consumer research is cognitively biased. Stern (1993,
p.562) argued that “incorporating feminist ideas into the research stream” would enhance the
roles of moods and feelings: “executional cues in ads are likely to evoke different moods
depending on the consumer’s gender”.
Holbrook, O’Shaughnessy and Bell (1990, pp.134-5) list no less than sixteen different non-
Cartesian research streams and ten methodologies. Collectively, “the experiential perspective”
is dominated by emotions and social interactions.
Advertising
Research into how advertising works, i.e. intermediate effects, is dominated by “persuasive
hierarchy” or “hierarchy of effects” models (Holbrook 1986; for reviews, see O’Shaughnessy
7
1992 pp.63-83; Vakratsas and Ambler 1996). In these, advertising information is mentally
processed sequentially in a series of stages, typically C Ö A Ö B (cognition Ö affect Ö
behavior). More recent variants (e.g. Bloom, Edell and Staelin 1994) have more complex
hierarchies, but they all presume sequential stages with cognitive processing. Neither
experience nor memory is usually involved.
Ehrenberg’s ATR model of advertising (1974) proposed that brand Awareness and Trial were
largely random at the individual level, with advertising serving to Reinforce behavior. In this
model, cognition plays little, if any, part. Others, e.g. Ray et al. 1973, Deighton 1984, Hoch
and Ha 1986, and Maloney 1990, have similarly concluded that product experience dominated
any persuasive effects of advertising. These conclusions, otherwise known as the “weak
theory” (Jones 1990) of advertising (persuasive hierarchy is the “strong”), begins with
awareness (C) but sees experience (E) as forming attitudes (A) rather than the advertising itself.
The role of advertising is the maintenance of awareness and attitudes, rather than necessarily
increasing them.
Holbrook, O’Shaughnessy and Bell (1990) synthesized reasons and emotions in consumer
behavior. More integrative than the linear (C-A-B) hierarchy of effects model, “the
consumption experience [is] a gestalt-like network of simultaneous interdependencies” as
proposed by Gazzaniga (1985). Importantly, Holbrook et al. argue that the hierarchy of
effects, and any other sequential model of how advertising works, must be wrong because the
8
cognitive and emotional components of decision-making should be balanced, in parallel,
avoiding either polarity. They reject causality as being the product of cognitive bias (see
below).
Maloney (1990) reviewed consumer psychology’s contribution to understanding advertising
and also dismissed the persuasive hierarchy (p.347). He suggested we should look to
neuroscience for answers but, meanwhile, suggested two integrative models of how advertising
works (figures 19.1 and 19.2). The former links impressions of advertising with impressions of
brand determined on both sides by the levels of (advertising/brand) involvement. The latter has
three stages of intermediate effects, all labelled “memories”, before moving to behavioral
effects (sales).
Vakratsas and Ambler (1996) collected the different theories of advertising into a single
taxonomy based on their structure: C (informative only), A (affective only), CA (persuasive
hierarchy), CEA (weak theory), CAEA (more complex hierarchies) and hierarchy free models,
of which there were few examples. They drew two main conclusions:
1. Models should incorporate C, E and A since the exclusion of one is likely to cause the over-
estimation of the others; and
2. The concept of a hierarchy of intermediate effects is not supported. It has just been
assumed.
9
The persuasive hierarchy (CA) groups of models which has dominated the literature, appears to
fail on both counts.
In some models of how advertising works, the cognitive component is minimized, e.g. the
peripheral varieties of ELM (Elaboration Likelihood Model - Petty and Cacioppo 1981).
Other models separate advertising into primarily cognitive or affective types, e.g. the FCB Grid
which has four cells: (high/low involvement) x (think/feel) (Vaughn 1980). The ELM
recognises that processing may be largely affective which could be interpreted as being similar
to classical conditioning (Pavlov 1927) in which the unconditioned stimulus is the pleasure
derived from purchase/usage (unconditioned response) of the brand. The ad becomes
associated with the pleasure so that the conditioned stimulus (the brand) leads to the
conditioned response (buy the brand). Practitioners have recognised the stimulus/response
nature of advertising (e.g. Bullmore 1991) in which advertising should be understood, not in
terms of the information being conveyed, but of the consumer response it is trying to achieve.
Thus an advertiser may disparage its own product but do so in a way that causes the consumer
to like the product more.
Similarly, the ATR model (Ehrenberg 1974) could be regarded as a form of operant, or
instrumental, conditioning (Skinner 1938) where the person’s buying behavior, and product
usage, becomes rewarded by a subsequent reinforcing event, i.e. advertising which provides
pleasure and also reassurance that the buyer behaved well. For example, seeing Kelloggs Corn
10
Flakes’ (KCF) advertising at the same time as eating the cereal (post-purchase) will, if both
experiences are enjoyable, reinforce the KCF buying habit. Belk (1985) in seeking to explain
the poor predictive power of buyers’ intentions suggested (p.9) “it seems likely that these [low
involvement] items are also repetitively purchased without express intentions to do so”.
As a reminder that people react differently to the same ad, Moore et al. (1995) showed that
more emotional people react more strongly to emotional appeals but equally to non-emotional
appeals. Gould (1991) distinguished “vital energy”, a consumer attribute presumably akin to
“animal spirits”, Descartes’ expression for lower order, non-cognitive, drivers of behavior
which was adopted by Keynes (1936) to explain the success of non-rational investment
decisions. Holbrook and Batra (1987) analyzed the emotional content of ads and concluded
that the three key dimensions, pleasure, arousal and domination, mediated the effects of ad
content on attitude towards the ad and thence, partially, attitude toward the brand.
Krugman (1985) recalled a 30 year earlier Advertising Research Foundation study into
advertising recognition and recall research. It was shown then, and by neuroscience since
(Squire 1995), that these are different forms of memory and should not be seen as two
strengths of the same function. Advertising research should use both measures, not just one.
Singh et al. (1988) showed that recognition was the more sensitive and discriminating.
11
Childers and Houston (1984) demonstrated that imagery (pictures) is an effective memory aid
in advertising measured as delayed recall. Rothschild and Hyun (1990) used EEG
(electroencephalograph) technology to show that TV ad recognition was increased when right
brain was employed initially but left hemisphere dominated during the following seconds. This
built on their previous (Rothschild et al. 1988) EEG study of hemispheric processing of
commercials. Bilateral processing was greatest for rational commercials and least for
emotional, with mixed appeal commercials between those extremes.
The concepts that emerge from this analysis are the importance of memory and affect, the
absence of a hierarchy of intermediate affects, the significance of the behavior feedback loop
into memory (experience) and the importance of context such as competitive activities and
whether the category is high or low involvement. Intermediate effects are seen as important
partly because advertising practitioners seek to measure them to predict consumer behavior.
There is some way to go. Lodish et al. (1995) concluded “4. It is unlikely that there is a strong
relationship between standard measures of T.V. commercial recall and persuasion for
established brands and the sales impact of the copy.”
Intermediate effects and neuroscience
In reviewing the marketing (and some psychology) literature on the intermediate effects of
advertising, I have sought to highlight the different perceptions of the relative importance of
cognition, affect and product experience (memory). We can now compare those streams with
12
the evidence from neuroscience and then use both to construct a fresh model of how
advertising works. Before that, I draw attention to the impact of cognitive bias on advertising
research and argue that it partially explains why cognition has received relatively more attention
than affect or memory.
Cognitive Bias
Practitioners have long (Allport 1935) recognized the importance of attitudes but have been
frustrated by poor (0 Ù 0.3) attitude behavior correlations (ABCs - Wicker 1969; Fazio and
Zanna 1979). Kraus’ (1995) meta-analysis of 88 non-advertising studies, however, found a
mean ABC of 0.38 which increases to more than 0.50 “if they are measured at corresponding
levels of specificity” (p.70).
Poor ABCs may be partially explained by Wilson et al. (1989) whose research indicates that
cognitive analysis of attitudes disrupt their reliability. For example, attitudes towards makes of
strawberry jam (perceived qualities) closely matched objective quality measures (Consumer
Reports) until the respondents were asked to explain their reasons. The cognitive disruption
of consumers’ reported feelings, however, did not change subsequent behavior. The
phenomenon is less true where the respondent is an expert in the field. Presumably, an expert
has spent many years questioning his/her own attitudes and synchronized the two. Thus no
disruption is involved in doing it again. Where the application of cognition to affect is a relative
novelty, however, disruption will be more pronounced.
13
Marketers and academics alike, in trying to understand the consumer’s mental process, are
themselves using processes that create cognitive bias. Market research survey methodology,
for example, typically invites respondents to think about, and perhaps explain their decision
processes, in ways that they would not ordinarily use.
When someone with a clipboard asks an opinion of Tide laundry detergent, one feels impelled
to provide a sensible answer, a reason, even if Tide had not previously received a nanosecond's
thought. The responses are then cognitively processed by the researcher and passed along to
an audience of practitioners, or academics, who apply their own cognitive faculties. A report
which is irrational or a-rational will be discounted, even though it may represent reality. At
each stage from consumer to research data collector, to research analyzer, to marketer,
cognitive faculties are used in order to make sense of, communicate, and thereby filter out,
non-cognitive elements.
Cognitive bias can arise in different ways: disruption of attitudes (Wilson et al. 1989), research
process (above) and the selection of research objects which are compatible with that process.
Computer reasoning is inevitably more suited to the analysis of items that can be expressed in
computer language just as a telescope is a poor listener.
14
One consequence of cognitive bias will be to underestimate the relative importance of other
mental factors. For an alternative perspective of cognition, affect and memory/experience, we
turn to the neuroscience literature.
Neuroscience
“The central tenet of modern neural science is that all behavior is a reflection of brain function”
(Kandel 1991a, p.5). Customer and marketer behavior may be reviewed in the context of
what neural science can tell us of what goes on in their “minds”. Some emotions, e.g. visceral,
and memories, e.g. hormonal conditioning, take place outside the physical brain but the
distinction need not concern us here.
Within the mind, we are concerned with three functions: memory, affect and cognition. The
brain has other functions which deal with receiving and transmitting information, bodily control
and genetic factors which will not be considered further even though Tesser (1993) argues that
some attitudes are moderated by heritability. In other words, our genes collectively predispose
us to like certain things, and thus perhaps brands, more than others. After reviewing the
implications of memory, I shall consider the interactions that are now known between feelings,
thinking and decision-making.
15
In essence, if we understand how our minds work, we may better understand the intermediate
effects of advertising both as inputs, as memory storage of advertising and experience and
determinants of behavior.
Memory
Memory can be categorized into four types (Kupfermann 1991a, p. 1002), see Figure 1:
Figure 1 here
“Procedural”, or “reflexive” or “implicit”, memory is not dependent on awareness, association,
consciousness or cognitive processes (Anderson 1983; Rose 1993). It records how to do
something. “Declarative”, or explicit, memory holds information about specific personal
experience, facts and events (Anderson 1983; Dudai 1989; Squire 1995). The word “memory”
here is therefore being used more broadly than its everyday use. Declarative memory in turn
dichotomises to “semantic” (meanings and associations, e.g. a red light means “stop”) and
“episodic” (what one did or said or what happened). Thus experience, through habits and
conditioning, “alters behavior nonconsciously without providing access to any memory
content” (Squires 1995, p.67). This may also be true for advertising.
Kupfermann (1991a, p.1003) summarizes the neural basis of memory with “just four
generalizations:
16
1. memory has stages and is continually changing,
2. long-term memory may be represented by physical (or plastic) changes in the brain,
3. the physical changes coding memory are localized in multiple regions throughout the
nervous system, and
4. reflexive and declarative memories may involve different neuronal circuits.”
We should elaborate each of these in turn.
Stages
The key separation, at least for declarative memory, is between short- and long-term. Short-
term storage may last “only for minutes at most” (Kupfermann 1991a, p.1003) or up to about
six hours (Rose 1993) before the memory is either transferred to long-term or lost. The
explanation of the different timings probably lies in some intermediate stages before memory
becomes permanent.
One implication for marketers is that a truly new concept requires considerable impact and/or
reinforcement to move through the stages to become a permanent memory. Frequent
repetition may be needed within the time limits of short-term memory. Existing memories,
however, may require very little refreshment. This is consistent with the advertising research
finding associated with Jones (1995) that one or two exposures per inter-purchase period may
be “enough”.
17
Memories are physical
This paper spends some time on the physical characteristics of memory in order to understand
how advertising effects, and experience, are stored. In the Anderson (1983, p.19) computer-
like model, for example, outside events are encoded (input) and performance generated
(output) to/from working memory which in turn feeds/is sourced by declarative memory
(facts) and procedural memory (programming rules). This is a useful model in some respects
but, as I shall show, differs significantly from current neuroscience. In particular, memory can
be moderated by affect with emotionally arousing films being better recalled than neutral
subjects (Cahill et al. 1996).
Memory holds everything that we have been taught or become accustomed to doing and all we
know about our needs, wants, attitudes and experiences. Unless changed by some fresh inputs,
it determines future behavior. Thus brand memory and brand equity are at least related
concepts. I will suggest that they are identical.
When two nerve cells (neurons) are connected, the point of contact is a synapse (Sherrington
1906; Ramón y Cajal 1911) - see Figure 2. Hebb (1949) deduced, since experimentally
confirmed, that the strengthening of the synapse, i.e. memory formation, required the near
simultaneous arrival of multiple stimuli. Figure 2 abbreviates the relative length of an axon
which is the fine thread which transmits information between parts of the brain (body). The
18
human brain has about 1011 neurons each of which has about 1,000 synapses, i.e. we have
about 1014 synaptic connections (Kandel 1991b, p.121). “A piece of the brain the size of a rice
grain contains, in addition to a million neurons, about twenty miles of axons.” (Stevens 1995,
p.37). And also ten billion synapses. Memory corresponds with the growth of new synapses.
One new synapse was thought to represent one unit of memory (Goelet et al. 1986) though this
is now thought to be an over-simplification (Rose 1996). Information is exchanged between
neurons by both chemical and electrical transmission, via axons, through these synapses.
Kandel (1991c, p.1014) suggests that “the long-term storage process appears to be a graded
extension of the short-term process” though this too is under review (Rose 1996).
Figure 2 about here
In this and many other respects, the biological brain is logically, as well as physically, very
different from the electronic computer misleadingly used as a metaphor (Churchland and
Sejnowski 1992; Lyons 1995). Humans remember meanings and routines, not data. Each time
we review a memory, we return it changed in some respect. A computer returns to store,
unless programmed otherwise, what it withdrew. Computers are not, yet anyway, affected by
concepts such as moods, feelings, and values.
Emotions moderate short-term memory for amnesic as well as healthy controls patients
(Hamann et al. 1997) although elderly individuals obtain less benefit from this phenomenon
19
(Nielson and Jensen 1994). As noted above, emotion also enhances long-term recall (Cahill et
al. 1996). Curiously perhaps, muscle tension also heightens retention (Nielson et al. 1996).
Multiple regions
Location is important to the marketer to the extent that if two functions, e.g. feelings and
decision-making, are identically located they may be closely connected whereas if they are
distant, they must be different.
Short- and long-term memories may be physically separate, and and declarative memories may
involve separate circuits (Kupfermann 1991a, p.1005), but we should not conclude that
particular memory types necessarily belong in particular areas. This is not yet established.
Squire (1995, p.73) concludes “memory is localized in the sense that different parts of the brain
store different aspects of information, but memory is distributed in the sense that multiple areas
of neocortex participate in representing even simple pieces of information.” The brain uses
parallel processing of information by association. Cueing one part of the memory can
reassemble the whole from its various locations. If some locations are blocked, or cut off
surgically, other locations may be employed (Kupfermann 1991a, p. 1005). This may be
limited to procedural memory as damage to certain parts of the brain can prevent new long-
term declarative memories being formed without hindering the recovery of existing memories.
20
Kupferman’s caution above that procedural and declarative memory “may” involve separate
circuits is not shared by Kandel and Hawkins (1995) who indicate (p.51) that they definitely
do. Procedural memory is slow to build.
Brain damage reports do not necessarily infer memory functions. The removal of a radio
component may cause the radio to squeal but that does not imply that the component’s
function is squeal suppression (Gregory 1961). Care with localization applies to all brain
functions, not just memory: “assigning functions to specific regions presents a problem , since
no part of the nervous system functions in the same way alone as it does in concert with other
parts” and “multiple areas of cortex are typically activated by even simple tasks” (Kupfermann
1991b). The brain is complexly integrated.
Procedural and declarative memories may involve different neuronal circuits
Figure 2 above shows the different forms of memory some of which seem to work
independently: “Some of the most compelling evidence for the newer [multiple] view has come
from findings that amnesiac patients, who are severely impaired on conventional memory tests
that assess recall and recognition of previously encountered material, are nevertheless fully
intact at many kinds of learning and memory” (Squire 1995, p.67). In other words, their
declarative memory had failed but their procedural had not. When patients were primed with
pictures of simple objects, amnesiac patients could also name them some days later faster than
they had originally. Squire’s conclusion echoes the EEG research noted above (Rothschild and
21
Hyun 1990) that an initial right brain, presemantic, visualisation is important for memory. That
the amnesiac patients had impaired declarative memory was confirmed from their inability to
identify which pictures they had seen before.
Memory implications
We have been considering memory because it holds any intermediate effects of advertising. In
particular:
• it holds everything that we have been trained, habituated, or conditioned, to do;
• it contains all we know about our needs, wants, and how well the choices we made in the
past then satisfied those needs and wants;
• unless changed by some fresh inputs, it determines future behavior;
• creating new memories make require different (amounts of) inputs from refreshing existing
ones in order to shift memory from short- to long-term;
• recall and recognition are separate phenomena, not different strengths of the same one; and
• separate memory systems, and especially declarative and procedural, appear to work in
parallel.
The physical basis of human memory, with its multiple circuits, forms of storage and
dependency on affect, throws light on brand equity, a much debated term. Srivastava and
Shocker’s (1991) definition will serve here: “a set of associations and behaviors on the part of a
brand's customers, channel members and parent corporation that permits the brand to earn
22
greater volume or greater margins than it could without the brand name and that gives a strong,
sustainable and differential advantage”. “Associations” implies conscious (declarative) memory
but “behaviors” imply that action has been taken, i.e. the sale has been made. A closer look at
the brand equity concept would regard it as a form of storage of marketing effects not yet
converted into performance. We have established (Kandel 1991a) brain function, memory, as a
predictor of behavior. The analysis above would indicate that “behaviors” should be replaced,
in a definition of brand equity, by those aspects of procedural memory which give rise to those
behaviors. Thus, brand equity should comprise intermediate stored effects, rather than realised,
and therefore transitory, effects.
Srivastava and Shocker consider customers (end users), channel members and employees
which could be generalized to the “relevant population”, i.e. those whose behaviors affect
brand performance. If we accept that brand equity is properly stored effects, rather than
results, and include both declarative and procedural memory, we can envision brand equity as
ΣM where M represents the brand memory of the relevant poulation.
In other words, the brand has an identity which accumulates meanings (awareness, associations
and attitudes) and habits (usage experience) which are stored in human memory. Given the
time lag between many marketing activities, such as advertising, and changed consumer
behavior, such as purchasing, brand equity can be seen as what is held on deposit, pending
market effects being realised. Thus brand equity exists as physical, albeit microscopic, parts of
23
the brain which we could in principle see, albeit not distinguish from other memories. One day
perhaps, marketers will be able to satisfy their accounting colleagues by directly measuring this
work in progress.
This discussion sets up the first leg of the model developed in this paper. M(emory) determines
our behavior in the absence of fresh A(ffect) or C(ognition), and also creates the context for the
fresh A or C, i.e. M⊃{A,C}. Inputs, such as advertising, are mediated by memory not least
because of the time lag between the perception of the advertising and the purchasing or
consumption behavior which it might affect. Direct response and point of sale advertising
aside, marketing activities can only affect behavior indirectly.
Affect and Cognition
As mentioned above, we are concerned with the physical location of brain functions because of
the inference of connectedness. In particular, the parts of the brain that deal with decision-
making and feelings coincide. Cognition may interact but it is more distant. It appears that
decision-making is more affective than cognitive (A⊃C). This may be compounded by the
proximity of feelings with social skills.
Cognitive functions are associated with the pre-frontal lobe (Figure 3).
Figure 3 about here
24
The hippocampus (so called because it looks like a sea-horse) seems to be involved in spatial
learning and, crucially, the conversion of short- to long-term memory (Rose 1993). LeDoux
(1992) distinguishes emotional memory from memory of emotion. The former is mediated by
the amygdala whereas the latter is mediated by the hippocampus. Primary emotions exist in the
primitive (central) brain (connected with the amygdala) whereas the more subtle feelings arising
from emotions and other (dis)pleasing effects of the environment and bodily states are located
(but see caution above) in the lower part of the neo-cortex (the ventro-medial section of the
frontal lobe - VMFL) (Damasio 1994).
Damasio (1994, p.145) distinguishes emotions (body state) from feelings which are the
experiences of that emotion “in juxtaposition to the mental images that initiated the cycle.”
He then (p.150) refers to three levels of feelings as being “Feelings of Basic Universal
Emotions, Feelings of Subtle Universal Emotions and Background Feelings” respectively. The
distinctions between emotions and feelings are not universally shared. For the purposes of this
paper, I shall maintain the same three levels described now as primary, or primitive or primal,
emotions, feelings and mood. Primary emotions (love, fear, hate etc.) may be in the brain or
elsewhere in the body but feelings in the sense used here are located in the VMFL, albeit linked,
since the brain is massively inter-connected. The third (weakest) level, mood, is likewise only
indirectly connected with the VMFL. “Feelings” here include concepts such as liking, warmth,
25
well-being, contentment and possibly humor. They certainly include group sensitivities such as
belonging and aptness.
In 1848, Phineas P. Gage suffered an iron rod through his brain (Damasio 1994) and not only
survived, but appeared to recover all his faculties. Soon after the accident, he was able to
converse normally. The damage was to the VMFL. In time, Gage was seen to have reduced
social skills and an inability to make decisions. He could no longer hold down a supervisory
job.
Damasio's main research source, code-named "Elliot", provided more immediate evidence that
damage to the VMFL did not impair reasoning, memory and communication skills but, like
Gage, Elliott had drastically reduced social skills, could not make decisions, nor maintain
employment. He was intellectually aware of the feelings he should have had, but did not.
Damasio’s conclusion was that decision-making is primarily associated with the part of the
brain that deals with feelings and social skills, not with the parts that deal with cognition. The
inference is that not only are advertisers likely to achieve better recall through use of affect but
also feelings are closer to the choice mechanism of the brain. We should, it appears, concur
with feminist writers (e.g. Hirschman 1993) and give affect precedence over cognition in
understanding decision-making.
26
The co-location of social skills and feelings in the VMFL with decision-making, has important
implications for advertising research. Many buying decisions have social aspects with friends,
office co-workers and families. The evolution of feelings in conjunction with these social
sensitivities, may be no accident. Over millions of years, our brains have evolved primarily to
ensure survival. Primitive emotions help us with immediate dangers, fear in the face of an
oncoming vehicle for instance, but the wider environment is immensely complex. The later
development of the brain, which includes the VMFL, helped us survive by being part of one or
more social groups. Mithen (1996, pp.110-111) suggests that the rapid increase of brain size
two million years ago was linked with the size of social groups. Language and social
intelligence developed from the need to replace physical grooming, which is limited to one
individual at a time, “and functioned to spread feelings of mutual content and well-being”.
Thus the feelings and sensitivities in the VMFL help us to integrate within our social groups.
This review of brain physiology leads to the conclusion that, for decision-making purposes,
A⊃C, and, joining that with the earlier memory implications, M⊃A⊃C, where M stands for
memory, A for affect or feelings, C for cognition and the ⊃ sign means “includes”, not in the
formal mathematical sense of set theory, but “sets the context for”. Memory, as we have seen,
includes both conscious memories and nonconscious habits and conditioning. More
specifically, advertising acts to enhance consumer decision-making in favour of the brand. Its
intermediate effects are stored as memories which tend, in turn, to be strengthened more by
affect than by cognition.
27
We must be wary of over-generalization here. Not all ads work the same way and different
people, e.g. the elderly, seem to be differentially impacted by A relative to C. The purpose of
the review thus far has not been to provide a blueprint for advertisers but to suggest that
memory and affect deserve prior consideration to cognition in marketing.
A Model of the Intermediate Effects of Advertising
Holbrook, O’Shaughnessy and Bell (1990, p137) integrated competing schools of thought into
“An Integrative Overview of the Consumption Experience” with three types of component:
reasons (thoughts, intention), emotions (wants, appreciation) and memory (habit,
reinforcement, experience). Usage reinforced the emotional components which, in turn, acted
on the rational and habit components which drove acquisition. At the same time, thoughts and
wants, looped with emotions, also moderated reasons.
This paper proposes a simplification of that analysis to a M ⊃ A ⊃ C (or “MAC”) model as
described above, i.e. memory, in the broad sense used here, is the primary driver of, or sets the
context for, decision making. None of us operates with total amnesia. Within memory, the
area of the brain concerned with (social) feelings is involved, but thinking (cognition) is not
essential. Clearly Damasio did not mean to suggest that VMFL lesion patients could make NO
decisions, or they would not even have been able to cross the road. He was dealing with those
28
where the brain is actively involved, specifically in changing a strategy. The thesis here is that
Damasio’s conclusions can be applied to the intermediate effects of advertising but we really
need to understand how advertising is processed when it is received, how the effects are stored
and how behavior is impacted.
Jones (1990) asks us to choose between the strong and weak (CA and CEA) models but it
seems more likely that advertising conditions us, depending on the circumstances, in both
classical and operant ways. New products will tend more to the former, persuasive, and
frequently purchased, low value, low involvement packaged goods more to the former. Clearly
one cannot reinforce a habit that does not yet exist but casting advertising in the role of building
and then refreshing brand synapses may help reconcile the two perspectives.
Marketers recognize, in their use of focus groups which usefully combine feelings with social
group dynamics, the importance of VMFL mental processing. Such research is not quantitative
(C) but those practitioners behind the one-way mirrors can empathize with the discussion
group (A).
Figure 4 sets out a conceptual model of how consumer brand display advertising is
hypothesized to work:
Figure 4 about here
29
At the input stage, the model recognizes that ads fight for attention with other stimuli and
competitive ads. Filters ensure that most potential perceptions do not engage our mental
faculties, but those that penetrate at all trigger memory. Within that, affect may be involved
and possibly cognition. The nesting shown in figure 4 implies significance, not sequence.
Attitudes and beliefs (memory) influence all stages of inbound processing; and information
supporting beliefs is more readily accepted and remembered (for overviews of the attitudes
literature see, inter alios, Johnson 1991, Eagly and Chaiken (1993), and Olson and Zanna
1993).
Thus long-term brand memory of usage, previous communications and associated material
(brand equity) is drawn down into short-term memory for affective and cognitive parallel and
interactive processing. Behavior is not directly involved except where the advertising and
brand-related behavior are simultaneous, e.g. direct response or point of sale advertising.
On the other hand, both before and after the advertising stimulus, behavior moderates
perceptual filters (buying a car makes one more likely to notice an ad for that car). Behavior is
also processed in short-term memory and thereby modifies brand equity.
30
Thus the MAC model incorporates both the strong and weak theories of how advertising
works by allowing for both persuasion and behavior reinforcement. It separates short- and
long-term memory effects and allows for competition and brand equity.
The MAC model adds affect and nonconscious memory to Anderson’s (1983) ACT* model
which also separates working from long-term declarative and procedural memories. It also has
similarities with Maloney’s (1990) “Hierarchy of Responses to Advertising” (Figure 19.2) but
eliminates the concept of hierarchy (as Maloney himself suggests). It seems much more likely
that intermediate effects are simultaneous. The model recognizes the following factors which
many extant models (Vakratsas and Ambler 1996) ignore:
• Brand advertising exists in the context both of direct and indirect competition for attention;
• Most of the messages do not penetrate our perceptual filters;
• Messages that do penetrate, operate, if they have any continuing effect at all, by changing
long-term memories, whether or not they have stimulated A and/or C;
• There is a time interval between advertising and subsequent decision making. Advertising
effects are stored as increased (relative to what would otherwise have been the case) brand
equity; and
• Behavior is a reflection of memory, affect and cognition which then loops back as feedback
to change/maintain awareness, associations, attitudes and conditioning.
31
Implications for managers
The MAC model supports the attention now given to brand equity, e.g. Aaker (1991, 1996);
Keller (1987). Direct response and point of sale advertising apart, advertising can only
indirectly affect sales. The brand equity asset at any point in time resides in human memory: If
advertising has any effect at all, it changes memory. In other words,
AdFX = ∆M = ∆ Brand equity
where AdFX is advertising effects and the changes are relative to what would otherwise (no
advertising) have been the case. Managers should recognize that memory is not a single entity
but has separate components, the main distinction being between declarative and procedural.
Their relative importance is likely to vary between brands, individuals and situations. We have,
however, established enough to suggest that managers be thoughtful in their use of market
research which
• measures declarative, not procedural, memory;
• disrupts affect and tends toward cognitive bias.
Whether the intermediate effects translate into behavioral changes depends on context as well
as advertising. Marketers seeking to measure the performance of their advertising, and/or their
agencies, may track both cognitive and affective brand memories and those of competing
32
brands, as best they can, with these caveats in mind. Advertising recall and recognition indicate
the penetration of perceptual filters, though not necessarily any change to brand memory.
The MAC model (figure 4) provides a benchmark to compare with the marketers’ own model
of how their advertising works. Target indicators should be determined in the context of the
likely level of involvement set by the product category, the competition, the balance of the
marketing mix, the stage of product life cycle (new versus refreshment of memory) and target
market. Both briefing agencies and then conducting pre- and post-measurement using this
consistent model should at least clarify communications.
Unfortunately, pre- and post-measurement does not identify change in the sense above (relative to
what otherwise would have happened). Split channels or other parallel testing may be possible. The
practical difficulties of quantifying intermediate effects are likely to maintain the emphasis on
behavioral measures for the time being.
The principal implication is the focus it brings to the measurement of brand equity. Inputs, such as
share of voice, and outputs, such as market share or relative price, can only be indirect measures.
Their value relative to direct measures, e.g. awareness, associations and attitudes, is an empirical
question each firm can determine for itself from analysis of available data. Kraus (1995) advises
that, as a general rule, ABCs (attitude behavior correlations) will tend to be higher when measured at
corresponding rates of specificity. For example, intention to purchase next week is likely to be more
predictive than broad brand liking.
33
Future research
Neuroscientific techniques to measure brand memory (brand equity) precisely are still beyond
both today’s science and marketing analysis. Directionally this may be the way to go, but in the
short-term the MAC model can be tested at each stage:
• The organization of perceptual filters. Figure 4, for example, shows feedback control from
behavior but from neither short-term processing nor long-term memory. Introspection
would indicate that we do not consciously control these filters. Some may consider it more
probable that the filters are determined by procedural memory and that behavior is mediated
thereby;
• Intermediate effects. It would be interesting to compare measures of declarative and
procedural memory after advertising but before purchase/consumption with those after
purchase/consumption but before the next flight of advertising;
• As noted above, the existing research comparing ads based on cognitive and affective
appeals is undeveloped (e.g. Rothschild et al. 1988). This area could be integrated with
single source data tracking the behavioral responses to split channel broadcasts of the two
types of commercial. The contextual factors noted above would also need to be recorded.
Such a holistic approach would be difficult to achieve for academics and firms who may
have already conducted such research, might not wish to release results due to their
commercial sensitivity;
34
• Consumer decisions could be categorized and tested within the MAC framework. Purely
routinized (nonconscious/barely conscious) choices are hypothesized to be driven by M
without “involvement” (Krugman 1965, 1967) which has both affective and cognitive
components. Following Petty and Cacioppo (1981), they could be separated into
elaboration (cognition) and arousal (affect) before replicating prior involvement studies, e.g.
Vaughn (1980), using this additional distinction. The objective would be to determine the
relative importance of M, A and C and the extent to which it is invariant across brands in a
category, across consumers and across usage situations. For example, the choice of ice-
cream may be peripherally elaborated when it is a routine family purchase but centrally when
it is for a dinner party;
• Decision making research may be extended from this paper in two directions: from
advertising to other forms of decisions and from the individual to small groups, e.g. families
where decision making is informal and empathetic bonds are strong. The size of the
decision making unit may not be as critical as the two latter conditions. It may even apply
to a single decision made by millions of people. In the 1996 US Presidential campaign,
advertising seemed to owe more to a MAC framework than anything Cartesian. The
research issue, therefore, is the extent to which MAC model applies not just to individuals
but to social groups.
35
Conclusions
The marketing literature has been developing in parallel streams. Pure cognitivists have had
little truck with relativism which is regarded as soft (e.g. Hunt 1992). Conversely, humanists
are impatient with what they consider mechanical models (e.g. Hirschman 1993). Neither are
wrong and perhaps neuroscience, which places hard measures on soft concepts, such as
memory and affect, may provide the bridge to bring these two streams together.
The two literatures together support the unlikelihood of any hierarchy of intermediate
advertising effects; they act simultaneously and interactively. Similarly, they support the idea
that advertising models should involve M, A and C; the elimination of any one is likely to
exaggerate the importance of the others. Advertising models should also include the context,
namely the likely level of involvement set by the product category, competitive advertising
effects, the balance of the marketing mix, the stage of product life cycle and target market.
Different forms of memory are affected differently by the combination of advertising and experience.
One should be cautious of choosing between persuasive and habituation models since it seems more
likely that both operate in parallel with their relative importance being determined by circumstance.
This is an argument for a holistic, non-reductionist, view of advertising effects whilst recognizing the
practical research difficulties involved.
36
The recognition that advertising directly affects brand equity and that brand equity resides in the
memories of the relevant population has implications for the briefing of agencies and the tracking of
performance. Both need to take place within the context noted above but also the specific goals of
the advertising. Advertising is only as good as the shift in brand equity, the collective brand memory
of the relevant population, it creates.
The growing interest in affective research may indicate shift from a Cartesian (purely cognitive)
to a Post-Cartesian understanding of marketing decision-making. In decision-making terms, I
have suggested that memory sets the context for affect which sets the context for, even
dominates, cognition (M⊃A⊃C). This is not to say rationality is unimportant. Where M and A
are neutral or conflicting, C is likely to be the deciding factor if involvement is high enough to
bring it into play. Otherwise choice is likely to be random.
Post-Cartesianism thus sees humans not as value-neutral, transactional, individual utility
maximizers but as sentient, experiential beings finding satisfaction in social groups. We need to
humanize our perception of decision making.
37
References
Aaker, David A., (1991), Managing Brand Equity, New York: Free Press.
----- " -------, (1996), Building Strong Brands, New York: Free Press.
Alba, Joseph W., J Wesley Hutchison, and John G. Lynch, Jr. (1991), “Memory and Decision Making.”
In Handbook of Consumer Behaviour, Eds. Thomas S. Robertson and Harold H. Kassarjian.
Englewood Cliffs, NJ: Prentice Hall, 1-49.
Allport, G.W. (1935), “Attitudes”, in The handbook of social psychology, C. Murchison (ed.)
Worcester, MA: Clark University Press.
Anderson, John R. (1983), The Architecture of Cognition, Cambridge, MA: Harvard University Press.
Batra, R. (1986), “Affective advertising: Role, processes and measurement”, in R.A. Peterson, W.D.
Hoyer and W.R. Wilson (eds.), The role of affect in consumer behavior, Lexington, MA, D.C. Heath,
53-85
Belk, Russell W. (1985) “Issues in the Intention-Behavior Discrepancy”, Research in Consumer
Behavior, 1, 1-34.
38
Bettman, James R. (1979), An Information Processing Theory of Consumer Choice, Reading, MA:
Addison-Wesley.
Bloom, Paul N., Julie Edell and Richard Staelin (1994), “Criteria for Assessing Research on the
Effects of Marketing Communications,” MSI Report, No. 94-123 (December).
Bullmore, Jeremy (1991), Behind the scenes in advertising, Henley-on-Thames: NTC Press.
Cahill, L., R.J. Haier, J. Fallon, M.T. Alkire, C. Tang, D. Keator, J. Wu, and J.L. McGaugh (1996),
“Amygdala activity at encoding correlated with long-term, free-recall of emotional information, ”
Proceedings of the National Academy of Sciences of the United States of America, 93(15), 8016-8021.
Calvin, William H. and George A. Ojeman (1994), Conversations with Neil’s Brain: The Neural
Nature of Thought and Language, Reading, MA: Addison Wesley.
Chaffee, S. H. and J. M. Macleod (1973), “Consumer Decisions and Information Use,” in Consumer
Behavior: Theoretical Sources, S. Ward and T. S. Robertson (eds.), Englewood Cliffs, NJ: Prentice
Hall, 385 - 415.
Childers, Terry L. and Michael J. Houston (1984), “Conditions for a Picture-Superiority Effect on
Consumer Memory”, Journal of Consumer Research, 11 (2, Sep), 643-654.
Churchland, P.S. and T.J. Sejnowski (1992), The Computational Brain, Cambridge, Mass.: MIT Press.
39
Csikszentmihalyi, Mihaly (1993), The Evolving Self, New York: Harper Perennial, 66-67.
Damasio, Antonio R. (1994), Descartes’ Error: Emotion, Reason and the Human Brain, London:
Papermac (Macmillan).
Deighton, John (1984), “The Interaction of Advertising and Evidence,” Journal of Consumer
Research, Vol. 11 (December), 763-70.
Dudai, Y (1989), The Neurobiology of Memory, Oxford: Oxford University Press.
Eagly, A. H. and Chaiken, S. (1993), The psychology of attitudes, Ft Worth, TX: Harcourt Brace
Jovanovich.
Ehrenberg, Andrew S.C. (1974), “Repetitive Advertising and the Consumer,” Journal of Advertising
Research, 14 (April), 25-34.
Engel, James F., Roger D. Blackwell and Paul W. Miniard (1986), Consumer Behavior. 5th Edition.
Chicago: Dryden.
Fazio, Russell H. and Mark P. Zanna (1979), “Direct Experience and Attitude-Behavior Consistency.”
In Advances in Experimental Social Psychology, L. Berkowitz (ed.), New York: Academic Press.
40
Gazzaniga, M.S. (1985), “The Social Brain”, Psychology Today, 19 (November), 29-38.
Goelet, P., V.F. Castellucci, S. Schacher and E. R. Kandel (1986), “The long and the short of long term
memory - a molecular framework”, Nature, 322, 419-22.
Gould, Stephen J. (1991), “The Self-Manipulation of My Pervasive, Perceived Vital Energy Through
Product Use: An Introspective-Praxis Perspective”, 18 (2, September), 194-207.
Green, Paul E. and Yoram Wind (1975), “New Way to Measure Consumers’ Judgements,” Harvard
Business Review, 53 (July-August), 107-17.
Green, Paul E. and V. Srinivasan (1990), “Conjoint Analysis in Marketing: New Developments With
Implications for Research and Practice,” Journal of Marketing, 54 (October), 3-19.
Gregory, R.L. (1961), “The brain as an engineering problem,” in Current Problems in Animal
Behavior, W.H. Thorpe and O.L. Zangwill (eds.), Cambridge: Cambridge University Press, 307-330.
Hamann, S.B., L. Cahill and L.R.Squire (1997), “Emotional perception and memory in amnesia,”
Neurophysiology, 11(1), 104-113.
Hebb, Donald O. (1949), The Organization of Behavior,New York: Wiley.
41
Hirschman, Elizabeth C. (1993), “Ideology in Consumer Research, 1980 and 1990: A Marxist and
Feminist Critique”, Journal of Consumer Research, 19 (March), 537-55.
Hoch, Stephen J. and Young-Won Ha (1986), “Consumer Learning: Advertising and the
Ambiguity of Product Experience,” Journal of Consumer Research, Vol. 13 (October), 221-33.
Holbrook, Morris B. (1986), “Emotion in the Consumption Experience: Toward a New Model of the
Human Consumer,” in The Role of Affect in Consumer Behavior: Emerging Theories and
Applications, R.A. Peterson, W.D. Hoyer and W.R. Wilson (eds.), Lexington, MA: D.C. Heath and
Company.
--------“-------- and Rajeev Batra (1987), “Assessing the Role of Emotions as Mediators of Consumer
Responses to Advertising”, Journal of Consumer Research, 14 (3, December), 404-420.
Holbrook, Morris B., John O’Shaughnessy and Stephen Bell (1990), “Actions and Reactions in the
Consumption Experience: The Complementary Roles of Reasons and Emotions in Consumer Behavior,”
in Research in Consumer Behavior, JAI Press, 131-163.
Howard, John A. (1977), Consumer Behavior: Application of Theory, New York: McGraw-Hill.
Howard, John A. and Jagdish N. Sheth (1969), The Theory of Buyer Behavior. New York: Wiley.
42
Hunt, Shelby D. (1992), “For Reason and Realism in Marketing,” Journal of Marketing, 56 (April), 89-
102.
Johnson, Blair T. (1991), “Insights About Attitudes: Meta-analytic Perspectives,” Journal of
Personality and Social Psychology, 17(3 June), 289-299.
Jones, John Philip (1990), “Advertising: Strong Force or Weak Force? Two views an ocean apart.”
International Journal of Advertising, 9(3), 233-246.
-----“-------- (1995), When Ads Work, New York: Lexington Books.
Kagan, Jerome (1989), Unstable Ideas: Temperament, Cognition and Self, Cambridge, MA: Harvard
University Press.
Kandel, Eric R. (1991a), “Brain and Behavior”, in Principles of Neural Science (3rd Edition) (Eds.:
Eric R. Kandel, James H. Schwartz and Thomas M. Jessel), Norwalk, Conn.: Appleton and Lange, 5-
17.
Kandel, Eric R. (1991b), “Elementary Interactions Between Neurons: Synaptic Transmission”, in
Principles of Neural Science (3rd Edition) (Eds.: Eric R. Kandel, James H. Schwartz and Thomas M.
Jessel), Norwalk, Conn.: Appleton and Lange, 121-2.
43
Kandel, Eric R. (1991c), “Cellular Mechanisms of Learning and the Biological Basis of Individuality”,
in Principles of Neural Science (3rd Edition) (Eds.: Eric R. Kandel, James H. Schwartz and Thomas M.
Jessel), Norwalk, Conn.: Appleton and Lange, 1009-31.
Kandel, Eric R. and Robert D. Hawkins (1995), “Neuronal Plasticity and Learning”, in Neuroscience,
Memory and Language, Decade of the Brain, Vol. 1, R. Broadwell (ed.). US Government Printing
Office, 45-58.
Kassarjian, Harold H. (1978), “Presidential Address, 1977: Anthropomorphism and Parsimony,” in
Advances in Consumer Research, Vol. 5, ed. H. Keith Hunt, Ann Arbor, MI: Association for
Consumer Research, xii-xiv.
Keller, Kevin L. (1987), “Conceptualizing, Measuring and Managing Brand Equity,” Journal of
Marketing, 57 (1, January), 1-22.
Keynes, John Maynard (1936), The General Theory of Employment, Interest and Money, London:
Macmillan, 161.
Kraus, Stephen J. (1995), “Attitudes and the Prediction of Behavior: A Meta-Analysis of the Empirical
Literature”, Journal of Personality and Social Psychology, 21 (1, January), 58-75.
Krugman, Herbert E. (1965), “The Impact of Television Advertising: Learning Without Involvement,”
Public Opinion Quarterly, 29 (3), 349-56.
44
-----“-------- (1967), “The Measurement of Advertising Involvement,” Public Opinion Quarterly, 30 (4),
583-96.
----“---------- (1985), “Point of View: Measuring Memory – An Industry Dilemma”, Journal of
Advertising Research, 25 (4, Aug-Sep), 49-51.
Kupfermann, Irving (1991a), “Learning and Memory”, in Principles of Neural Science (3rd Edition)
(Eds.: Eric R. Kandel, James H. Schwartz and Thomas M. Jessel), Norwalk, Conn.: Appleton and
Lange, 997-1008.
Kupfermann, Irving (1991b), “Localization of Higher Cognitive and Affective Functions: The
Association Cortices”, in Principles of Neural Science (3rd Edition) (Eds.: Eric R. Kandel, James H.
Schwartz and Thomas M. Jessel), Norwalk, Conn.: Appleton and Lange, 823-838.
LeDoux, J.E. (1992), “Emotion and the amygdala”, in The Amygdala, J.P.Aggleton (ed.), New York:
Wiley-Liss.
Lodish, Leonard M., Magid Abraham, Stuart Kalmenson, Jeanne Livelsberger, Beth Lubetkin, Bruce
Richardson, Mary Ellen Stevens (1995), “How Advertising Works: A Meta-Analysis of 389 Real World
Split Cable TV Advertising Experiments,” Journal of Marketing Research, 32 (May), 125-139.
Lyons, William (1995), Approaches to Intentionality, Oxford: Clarendon Press.
45
Maloney, John C. (1990), “Consumer Psychology’s Potential Contribution to Social Science”, in
Emotion in Advertising, Stuart J. Agres, Julie A. Edell and Tony M. Dubitsky (eds.), Westport, CT:
Quorum Books, 329-367.
Mithen, Steven (1996), The Prehistory of the Mind, London: Thames and Hudson.
Moore, David J., William D. Harris and Hong C Chen (1995), “Affect intensity: An individual
difference response to advertising appeals”, Journal of Consumer Research, 22 (2, September), 154-
164.
Nedungadi, P. (1990), “Recall and Consumer Consideration Sets: Influencing Choice without Altering
Brand Evaluations”, Journal of Consumer Research 17 (December), 263-76.
Nielson, K.A. and R.A. Jensen (1994), “Beta-adrenergic-receptor antagonist antihypertensive
medications impair arousal-induced modulation of working-memory in elderly humans,” Behavioral and
Neural Biology, 62(3), 190-200.
Nielson, K.A. R.C. Radtke, and R.A. Jensen (1996), “Arousal-induced modulation of memory storage
processes in humans,” Neurobiology of Learning and Memory, 66(2), 133-142.
Olshavsky, Richard W. and Donald H. Granbois (1979), “Consumer Decision Making - Fact or
Fiction?” Journal of Consumer Research, 6 (September), 93-99.
46
Olson, James M. and Mark P. Zanna (1993), “Attitudes and Attitude Change,” Annual Review of
Psychology, 44, 117-154.
O’Shaughnessy, John (1992), Explaining Buyer Behavior, Oxford: Oxford University Press.
Pavlov, I.P. (1927), Conditioned Reflexes: An Investigation of the Physiological Activity of the
Cerebral Cortex, (translator: G.V. Anrep), Oxford: Oxford University Press.
Petty, R. E. and J. T. Cacioppo (1981), Attitudes and Persuasion: Classic and Contemporary
Approaches, Dubuque, IA: Wm. C. Brown.
Pham, Michel Tuan (1996), “Feeling like consuming: Affect as information in consumer decision
making”, Working paper, Columbia Business School.
Ramón y Cajal, S. (1911), Histologie du Système Nerveux de l’Homme & des Vertébrés, Vol. 2,
(republished 1955), Madrid: Instituto Ramón y Cajal.
Ray, Michael L., Alan G. Sawyer, M.L. Rothschild, R.M. Heeler, E.C. Strong and J.B. Reed
(1973), “Marketing Communications and the Hierarchy-of-Effects,” in New Models for Mass
Communication Research, P. Clarke (ed.), Beverly Hills, CA: Sage Publishing, 147-176.
47
Robertson, Thomas, (1976), “Low-Commitment Consumer Behavior,” Journal of Advertising
Research, 16 (2), April, 19-24.
Rose, Steven P.R. (1993), The Making of Memory, London: Bantam Books.
--------”----------- (1996), informal discussion with the author.
Rothschild, Michael L. and Yong J. Hyun (1990), “Predicting Memory for Components of TV
Commercials from EEG”, Journal of Consumer Research, 16 (4, Mar), 472-478.
-----“-----------, Byron Reeves, Esther Thorson, Robert Goldstein (1988), “Hemisperically Lateralized
EEG as a Response to Television Commercials”, Journal of Consumer Research, 15 (2, September),
185-198.
Sherrington, C. (1906), The Integrative Action of the Nervous System, (2nd edition, 1947) New Haven:
Yale University Press.
Singh, Surendra N., Michael L. Rothschild and Gilbert A. Churchill, Jr. (1988), “Recognition Versus
Recall as Measures of Televaision Commercial Forgetting”, Journal of Marketing Research, 25 (1,
Feb), 72-80.
Skinner, B. F. (1938), The Behavior of Organisms: An Experimental Analysis, New York: Appleton-
Century-Crofts.
48
Squire, Larry R. (1995), “Memory and Brain Systems”, Neuroscience, Memory and Language, Decade
of the Brain, Vol. 1, R. Broadwell (ed.). US Government Printing Office, 59-75.
-------“----------- and M. Zola-Morgan (1991), “The medial temporal lobe memory system”, Science 253
(September 20), 1380-86.
Srivastava, Rajendra K. and Allan D. Shocker (1991), "Brand Equity: A Perspective on its Meaning and
Measurement", Working Paper No. 91-124, Cambridge, Mass: Marketing Science Institute.
Srull, Thomas K. (1990), “Individual Responses to Advertising: Mood and Its Effects from an
Information Processing Perspective”, in Emotion in Advertising, Stuart J. Agres, Julie A. Edell and
Tony M. Dubitsky (eds.), Westport, CT: Quorum Books, 35-52.
Stern, Barbara B. (1993), “Feminist Literary Criticisms and the Deconstruction of Ads: A Postmodern
View of Advertising and Consumer Responses,” Journal of Consumer Research, 19 (March), 556-566.
Stevens, Charles F. (1995), “The Cellular Basis for Learning and Memory” in Neuroscience, Memory,
and Language, Vol. 1, Richard D. Broadwell (ed.), Washington, DC: Library of Congress, 37-44.
Strong, Jr, E.K. (1925), “Theories of Selling,” The Journal of Applied Psychology, 9, 75-86.
49
Tesser, Abraham (1993), “The importance of heritability in psychological research: The case of
attitudes,” Psychological Review, 100, (1), p.129-142
Turner, E.S. (1952), The Shocking History of Advertising, London: Michael Joseph.
Vakratsas, Demetrios and Tim Ambler (1996), “Advertising Effects: A Taxonomy and Review of
Concepts, Methods and Results from the Academic Literature”, Marketing Science Institute Working
Paper 96-120, Cambridge, Mass.
Vaughn, Richard (1980), “How Advertising Works: a Planning Model,” Journal of Advertising
Research, 20 (5), 27-33.
Wicker, A.W. (1969), “Attitudes Versus Action: The Relationship of Verbal and Overt Behavioral
Responses to Attitude Objects.” Journal of Social Issues, .25(4), 41-78.
Wilson, Timothy D., Dana S. Dunn, Dolores Kraft and Douglas J. Lisle (1989), “Introspection, Attitude
Change, and Attitude-Behavior Consistency: The disruptive effects of explaining why we feel the way
we do”, Advances in Experimental Social Psychology, Academic Press, 22, 287-343,
Zajonc, R.B. (1980), “Feeling and Thinking,” American Psychologist, 35(2), 151-175.
A Taxonomy of Memory
Memory
Reflexive/procedural/implicit
Declarative/explicit
Figure 1
Nonassociativelearning
Classicalconditioning
Primin g
Skills and habits
Events
FactsAdapted from: Squire & Zola-Morgan 1991
Figure 2 - Synaptic Connections
Synapse
Neuron
Source:Kandel1991a,p.19
Figure 3
MAc: A Conceptual Model of How Advertising Works
Ads
Perceptualfilters
Memory
Affect
Cognition
Long-termmemory:
brandequity
Behavior
Competitors for attention
Figure 4