50 television 2011 | international key facts

Brainwaves: what neuroscience can teach us aBout the Power of television

David Brennan and the team at thinkbox in the

uK have been at the forefront of research that

explores our understanding of the Power of tv.

their latest project covers arguably the most

important subject of all, how tv advertising affects

the brain.

We have learned more about how the human brain works in the

past ten years than we had learned in the previous couple of

centuries, driven in large part by advances in neuroscience.

We are now able to peek inside the human brain – even when

it is engaged in everyday activities – and work out what is going

on at an increasingly accurate level. Most importantly, it means

we don’t have to ask people to tell us what is going on in their

heads, a research approach that is often inaccurate and sub-

ject to misrepresentation. This is especially important when we

want to conduct research on the effects of television, which

often affects us on an emotional, implicit and subconscious

level. So, rather than asking people about these effects after

Pictures of SST in action Pictures of fMRI in action

Validated academically and clinically over 15 years, used commercially since 2005

SST benefits• Portable • Contextual sensitivity • Lowers cost, larger sample sizes• Time-based analysis

Glossary

SST: Solid state topography

LTME: Long Term memory encoding

fMRI: Functional magnetic resource imaging

EEG: Electroencephalography

Allows 3D visualisation of deep brain activity: reward, memory encoding, craving, etc.2,000+ benchmarking datasetsUsed commercially since 1999

fMRI benefits:• Encompasses all brain activity• Exact locations, directions and cause of responses• Depth and accuracy

Steady State Topography (SST) with Neuro Insight

Functional Magnetic Resonance Imaging (fMRI)

television 2011 | international key facts 51

the event (when they often forget, or

weren’t aware in the first place) or during

the viewing experience (which then inter-

feres with the “natural” state of viewing),

we can understand much more about

them by looking at which specific parts

of the brain are most active when they

are watching TV and what experiences

are associated with that activity.

There are two main ways of measuring

what is going on in the brain. Functional

magnetic resource imaging (fMRI) pro-

vides a three-dimensional image of

blood flow in the brain, allowing us to de-

termine which parts of the brain are most

active at specific times. It is highly accu-

rate and can measure deep brain activi-

ty, but the disadvantages are that sub-

jects must lie completely still in a full

body scanner and – because blood flow

changes are not instantaneous – it can-

not measure immediate changes in ac-

tivity very accurately. On the other hand,

electroencephalography (EEG) meas-

ures electrical impulses in the brain,

which change very quickly, and can be

measured via a small cap on the sub-

ject’s head, which frees them to do day-

to-day activities in a more natural setting,

but can only measure activity in the sur-

face areas of the brain.

We decided we needed to use both

techniques, so we could understand

how the brain was processing TV com-

mercials on a second-by-second basis,

in a natural setting, while also under-

standing which specific parts of the brain

were active across the whole commer-

cial. This would provide us with both the

depth and breadth of the viewers’ expe-

rience when they were watching TV.

We recruited seventeen major advertis-

ers who, between them, submitted more

than twenty-five pieces of content to be

tested. Some were tested via fMRI, oth-

ers via solid state topography (SST, a

superior form of EEG). We set up the re-

search so that we could measure the dif-

ferences in brain activity when people

were watching TV compared to when

they were interacting online (via a guided

online journey, taking in web browsing,

search and on-demand TV content). We

also wanted to look at how TV primes

online experience, and vice versa, as

this priming effect had been evident in

several of our earlier studies looking at

the relationship between TV and online.

The results highlighted just how powerful

TV ads can be in engaging us in their

content and transferring that content into

our long-term memory. They also taught

us about the importance of media con-

text; how the placement of an ad can turbocharge the cre-ative impact.

the power of tv creativity

The SST part of the study allowed us to

look at how TV commercials affect five

main parts of the brain, associated with

emotional intensity, emotional direction

(attraction or withdrawal), attention, en-

gagement and long-term memory en-

coding.

Previous studies have demonstrated that

it is levels of long-term memory encoding

(LTME) – especially for those parts of the

ad where the brand is featured – that has

the greatest effect on whether or not we

are likely to purchase that brand. The

challenge for creative agencies is in gen-

erating high levels of LTME and ensuring

that the brand or product is featured in

exactly the right places.

Our study demonstrated quite conclu-

sively that it is not attention that creates

long-term memory encoding, but en-

gagement. The neuroscientists define en-

gagement as “a sense of immersion in an

activity caused by personal relevance”,

Emotional direction (attraction or withdrawal),

attention, engagement and

long-term memory encoding

A research first!

SST – the breadthTIME SERIES ANALySIS:

the critical points at which the brain is being influenced

• 120 respondents• 60 TV first / 60 online first

fMRI – the depth FuLL bRAIN ANALySIS:

what is going on in the brain duringthese time-points and why?

• 20 respondents• TV only

it is not attention that

creates long-term

memory encoding, but

engagement: “a sense of

immersion in an activity caused

by personal relevance”

52 television 2011 | international key facts

which is just about the best definition of engagement I have

heard to date.

Across all of our measured advertisements, the correlation be-

tween levels of engagement and long-term memory encoding

was well over 70%. Engagement itself is influenced by emo-

tional intensity, so it is not surprising that this also correlated

strongly with LTME. It is interesting that visual attention, in either

left-brain (detailed) or right-brain (global) hemispheres had no

relationship whatsoever with either engagement or LTME. This

supports theories such as Heath’s theory of low attention pro-

cessing, which suggests that attention can actually reduce an

ad’s ability to create associations and alter perceptions.

One other measure also failed to correlate with LTME: that of

emotional direction (like/dislike or attract/repel). Normally one

would expect this to correlate with engagement and LTME, with

a higher level of liking or attraction relating to higher levels of

engagement, but again we saw no relationship. I think this is

due to a part of the creative process that we can’t define or

measure so easily: the narrative power of an ad to tell a com-

pelling story in a matter of seconds. All great stories have ele-

ments of conflict or challenge and without a degree of “nega-

tive” emotion it would be

impossible for the story to

create enough of a sense of

intrigue or involvement. One

of the most powerful ele-

ments of neuroscience is

that one can see the shape

of the story taking place in

people’s heads – the stronger

the storytelling, the more pro-

nounced the peaks and troughs

of engagement and memory encoding!

Of course, it is all well and good having high peaks of LTME,

but it is not much use if the brand is not present when those

peaks take place, and our analysis from this study demonstrat-

ed quite clearly one of the challenges in making this happen.

We know that the brain “chunks” information into manageable

pieces, and when it feels a “chunk” of the storyline has been

told, it will go into shutdown for a second or so while it pro-

cesses the information it has received. This is known as “con-

ceptual closure”, and we saw many examples in our study. The

problem is that it is often the point in the advertisement when

Levels of correlation between SST measures

Engagement(L=.73, R=.71)

Attract/Repel LTMEVisual

attention

Emotionalintensity(L=.58, R=.62)

Liked/Disliked

Mental effort

Personal identification

Strength of emotion

Engagementand strength of emotion strongly correlated with LTME

NO correlation with visual attention!

No significance

Highly significant

Source: Neuro Insight (SST)

the importance of creativity

in tv advertising

television 2011 | international key facts 53

the brand or pack shot comes into view,

just after the resolution or the “punch-

line”, but this is often the point in the ad

when the brain is least receptive to new

information. We were able to advise a

work creatively. Things like the power of

music to drive emotions, engagement

and attention was seen a number of

times, and those ads that contained mu-

sic performed significantly better on

those parts of the brain than those that

didn’t. We were able to measure the abil-

ity of our mirror neurons to respond well

to smiling faces and positive emotions

expressed within the ads we tested;

there were some clear examples of emo-

tion and engagement following the emo-

tions expressed by the characters in the

ads we analysed. And we were able to

use the fMRI analyses to see the impor-

tance of TV ads in creatively engaging

the parts of the brain responsible for pur-

chase decision-making or empathy, both

of which have a huge influence on how our

brains respond to a piece of advertising.

What all of this work demonstrated was

the importance of creativity in TV adver-

tising. The central idea, and the way it is

structured to tell a story, can have a huge

impact on how well an ad works. We

provided irrefutable evidence of that ef-

fect in a study we commissioned along-

side our neuroscience work. We ana-

lysed the IPA databank of entries to the

annual Advertising Effectiveness Awards

across the last eight years and com-

pared the performance of those that had

won major creative awards against those

that hadn’t. Although the creatively

awarded campaigns generally spent sig-

nificantly less on TV airtime, they outper-

formed the non-creatively awarded cam-

paigns on virtually every measure. More

importantly, they were able to drive mar-

ket share far more efficiently – in fact,

they drove share eleven times more effi-

ciently per pound spent than non-crea-

tively awarded campaigns. That is why it

is so important that the creative process

is better understood, and techniques like

neuroscience are perfectly placed to

provide that understanding.

number of our participating advertisers

on how to make simple editing changes

to overcome this problem.

We were also able to see some clear ex-

amples of what makes a TV commercial

TV and Internet Neuro-States0.9

0.8

0.7

0.6

0.5

0.4

0.3Mean

SSTEngagement

TV Internet

MemoryRight

MemoryLeft

VisualAttention Left

VisualAttention Right

Source: Neuro-Insight (SST) UK

Differences in brain activity – TV vs onlineTV associated with much higher engagement and memory encod-ing, but lower levels of visual attention – where the Internet was dominant (consistent across different types of online experience).

television 2011 | international key facts54

the importance of media context

Although the power of good creative

planning and execution has always been

known to affect how well a TV campaign

works, the power of media placement

has always been more difficult to evalu-

ate. Our neuroscience work demonstrat-

ed that where the advertising appears

can also have a profound effect on the

campaign’s success.

We specifically wanted to investigate the

differences between television viewing

and online activity, as our previous re-

search had suggested they played a

strong complementary role. We looked

at brain activity when people watched a

TV programme – complete with com-

mercial breaks featuring our participating

brands – compared to a range of online

activities, such as search, web browsing

and on-demand TV.

Across all of the commercials and online

ad formats that we tested, TV viewing

generated much higher levels of emo-

tional intensity, engagement (almost 30%

higher), detailed long-term memory (al-

most 30% higher) and global long-term

memory (more than 50% higher). As

mentioned earlier, these are the main

predictors of future brand purchase.

Online did generate more brain activity

for both left-brain and right-brain visual

attention (20% to 30% higher than for

TV). This is not surprising, considering

online activity tends to be more task-led,

and online activity requires active deci-

sion-making, which in turn requires at-

tention. What this means is that the high-

er engagement and memory encoding

of TV, mixed with the higher attention

levels online, produce all of the elements

needed for both brand building and re-

sponse generation.

The art of good media planning is not

just about placement but also about tim-

ing. Many of our previous studies had

previously identified a priming effect from

TV to online; if a commercial had already

been seen on TV, then the online pres-

ence of that brand would stand out far

more. When we conducted our study

with the IAB in 2007, awareness of online

display more than doubled if the re-

spondents had seen the TV work first.

We wanted to investigate whether this

phenomenon was related to brain activi-

ty, as well as to test whether it was a

one-way relationship or whether online

could prime TV as well.

We designed this study to investigate the

priming effect in depth. Half of our SST

sample saw the TV programme and ads

first, followed by the online activities; the

other half did it the other way round.

When we compared brain activity levels

across both activities, it became clear that

TV primes online by a significant amount.

Across all of the commercials we tested,

on average levels of engage-ment and LTME for online was much greater if re-spondents had seen the TV commercials for the featured brands first. It was just over 20%

higher for engagement (and a similar

level for emotional intensity) and an aver-

age 30–40% higher for LTME. We were

also able to map out the different priming

effect on different forms of online activity;

as expected, it was pre-rolls (video ads

in on-demand TV programming) where

the effect was the greatest, although we

also observed the effect across web

browsing and banner advertising. This

absolutely supports our earlier work in

this area and shows the added boost TV

advertising provides for online activity.

The effect was not, however, seen in re-

Internet scores without & with TV ad priming

0.9

0.8

0.7

0.6

0.5

0.4

SSTEngagement

Internet 1st TV 1st

MemoryRight

MemoryLeft

VisualAttention Left

VisualAttention Right

Source: Neuro-Insight (SST) UK

TV ads “prime” associated web activity processingSST found that memory and engagement online is enhanced when TV ads were viewed before associated web activity

55television 2011 | international key facts

verse; if respondents

had seen the online ac-

tivity (search, video ads

or website) first, they were

not more responsive to

the TV advertising.

We were also able to

identify some other media effects from

the neuroscience. The fMRI analysis

showed the importance of congruence

– TV commercials that were congruent

with the surrounding programming per-

formed significantly better on scores re-

lating to emotional saliency and memory

encoding. We were able to demonstrate

that ads perform better on average in

programmes which are more engaging.

We were also able to demonstrate that

ads which appear in the early part of the

break tended to perform much better

than those that appeared later. These

are all areas in which we would like to

conduct more work, but the neurosci-

ence uncovered media

effects well above what

more traditional re-

search methods have

ever been able to dem-

onstrate.

conclusions

Neuroscience as a media research tool

is still finding its feet, and there is a lot of

work to do in order to “calibrate” it with

actual behaviour. However, much of TV’s

impact is on the emotional, subcon-

scious and implicit mind, and the results

of this study (as with many others) sug-

gests there is a great deal more going on

when people are exposed to TV adver-

tising than those same people are able

to report or articulate for themselves.

What we also found with this study was

that our participating advertisers learned

far more about how their ads are working

than they could normally get from tradi-

tional questionnaire research. It opened

up discussions and insights into how

the creative process worked and where

it could be improved that are beyond

the scope of most creative pre-testing

research.

Finally, it taught us that TV advertis-ing works, often at a very deep level.

It is not just about messages and awareness, it is about creating emotional associa-tions, telling stories, influ-encing the way we feel about brands. That is why television con-

sistently comes out top in major effec-

tiveness studies: it’s all in the mind! ■

David Brennanresearch & strategy Director,

thinkbox

david.brennan@thinkbox.tv

Ads performbetter on average

in programmes which are more

engaging.

Attention

SST: TV commercials affect parts of the brain, associated with:

Long Term Memory Encoding

Engagement

Emotional Direction(attraction or withdrawal)

Emotional Intensity