Statute of imitations!

10 August 2015 / Tim Holmes

I don’t normally discuss my private life on this blog, but it’s already August and I am preparing to head out to the European Conference on Eye-Movements (ECEM) in Vienna, Austria, which for me can mean only thing. No, not that it’s time to dig out the Sound of Music on Blu-ray, although now I come to think of it… it can only be a matter of days before my mother initiates discussions about Christmas – so I thought I’d start this blog with a true story! Last October I got one of those panicky texts from my mum that start around this time and arrive pretty much once a week throughout the run-up to Christmas (or in her world, the second half of the year)! The text concerned a shopping trip for some chocolate liqueurs, one of the five basic festive food groups in our house, and the fact that my mum, in her mid-70’s, had arrived home only to discover that she hadn’t actually purchased the Baileys truffles she wanted, but some unknown brand called Kings instead. Christmas was instantly cancelled! I suggested to my mum that she hunt down the real Baileys product she wanted and then send me a picture of each of them, which she did (see below) – apologies for the quality of the images, but her phone’s not up to much and her eyesight is not exactly 20:20 either! Now, of course this relates to a previous blog I wrote on copycat brands, but a couple of things struck me about this example. Firstly, the packs are actually quite different, but the Kings logo clearly mimics the Baileys one and, more importantly, the bottle image kind of looks like the old style Baileys bottles – so I think we can safely call this a copycat. Secondly, I’m not sure I would have been fooled by this one, and I can’t help but wonder how much of that might come down to my mum’s failing vision, although of course, I cannot account for her attention at the time of purchase! Well, it turns out that we are ALL susceptible to such effects, but degraded visual acuity will certainly increase the likelihood of mistakes being made. So today I want to tell you about some preliminary results I am presenting at ECEM next week which explore this further from a visual search perspective.

Before I share the results, let’s start with a bit of theory about visual search – look for the green T in the three images below and notice which is the easiest and which takes you the longest.

The literature tells us that the two images on the left should be easiest and there shouldn’t be much difference between them. In the first image you find the T so quickly because you are able to dismiss (or deselect) all the other letters because they are not shaped like a letter T and, importantly it turns out, they are all shaped the same. In the second image you are able to use the colour to perform a similar deselection and, because the colour difference is slightly more salient, it might have seemed a little easier – in fact, studies show that there is little difference between the two, unless you’re one of the 7-10% of the population who are red/green colour blind of course!   The right most image should have been the hardest and that is because the other letters (distracters) all share an attribute with the green T (target), so now you can’t use either the colour or the shape to deselect since you would end up deselecting your target. The consequence of this is that your time to locate the target is much longer and if you’d only been shown the images for a limited amount of time you might not have found the target at all. Theories of visual search abound, and we can debate those another day, for now all we need to know is that the effects I’ve just shown you are highly replicable and have mostly been investigated using stimuli like those above.

One last bit of theory which I mentioned in the previous blog, but I’ll reintroduce here, is the effect of cognitive biases such as the availability heuristic. Put simply, this basically means we tend to make decisions based on a combination of prior experience and available current information. What this means in vision terms is that, depending on the angle and distance we’re viewing something from, as well as the degree of correction of any visual impairment we might have (fellow glasses wearers can take comfort in the fact that around two thirds of the UK population require some kind of correction), we are susceptible to incorrect decision making because we frequently don’t have ALL the necessary information available to us at the time we need to make a decision, so we end up playing a bit of a numbers game based on our past experience. To illustrate just how fallible we are, in the study I’m about to report we showed participants lots of products on supermarket shelves and when they’d finished the experiment we showed them a list of items and asked them to tick only the ones they’d seen. Of course, being sneaky neuroscientists we included some items on the list that they might have expected to be on our shelves but which weren’t actually there. 17% of our participants, yes you read that correctly, identified products they had not seen only a few minutes after they had…er…not seen them! So much for surveys huh?! The point here is that they identified these products because they expected them to be there and so when recalling the shelves they were actually remembered as being there.

OK, so let me quickly tell you the set-up and then we can get to those rather interesting results. 28 participants were shown 64 supermarket shelf mages – everyone saw all 64 images, the sequence was randomised and counterbalanced to control for possible learning, position and fatigue effects, and they were presented in blocks of 16 with short breaks in-between. No two shelf layouts were the same, although the products were, meaning that the previous location of a product was no clue as to its location on any other shelf. Products were horizontally blocked, as in supermarkets, and were presented life-size on (the world’s largest touch) screen at the Shopper Science Lab Research facility in Brentford, West London. For this study we worked with the 8 brand/copycat pairs shown in our copycat advertisement at the top of the post, and each shelf represented one of four conditions: lead brand present, copycat brand present, both lead and copycat brands present; neither lead nor copycat present – a useful control condition, as you will soon see.

These conditions meant that there was a 50:50 chance that the lead brand was present, and the same for the copycat. We also wanted to explore the performance of these packages when the visual signal was degraded, so we had four image type conditions: sharp focus/full colour, blurred/full colour, sharp focus/desaturated (or black and white to you and me) and blurred/desaturated to see whether colour and/or clarity contribute to the search and decide parts of the task differently. Blurring of the images also tells us something about the susceptibility to erroneous decision making introduced by a range of visual disorders such as cataracts, macular degeneration, diabetic retinopathy, astigmatism and simply uncorrected short and long sightedness. In all cases our products were presented as part of a planogram (a well-stocked shelf of blocked products from the same category), and participants were told to look for the lead brand and indicate when they had found it by pushing a button. We were of course eye-tracking them throughout and timing the button presses.

So what did we find? Well, I could show you a whole load of heat-maps but they simply show our participants looking at both the lead and copycat brands and don’t really explain what’s going on – this is a yet another great example of why heat-map based analysis is so limiting, so instead I’ll go to the data! Our participants had to make a decision about whether or not the lead brand was present. In order to do this accurately they had to search the shelf until they found it or search the shelf until they had decided it was NOT present. This search process is the first clue that copycat brands directly affect a shoppers’ ability to make the correct decision. In our study the average time to locate the lead brand when present (time to their first fixation longer than 250ms) was 3.12 secs when the copycat wasn’t on the same shelf and 3.31 secs when the copycat was on the same shelf. Not much of a difference, but this slowing down of time to locate the target is even more interesting when you consider that time to locate the copycat brand was 3.51 secs when the lead brand wasn’t on the shelf but 2.57 secs when the lead brand was present. In other words when both products appear on the shelf together the copycat brand is typically located FIRST and by nearly a full second! In the real world where shoppers are surrounded by distractions, under time pressure to make decisions and also looking to save money, the copycats clearly are stealing an opportunity to influence the purchase decision before shoppers even look at the lead brand. Blurring images didn’t affect this much, but across all conditions the time to locate increased when colour was taken out of the images, confirming that colour is an important brand attribute that can be used by the shopper to optimise the search process.

What about the decision making? Well first of all, our participants were certainly human, and made mistakes even when neither the lead nor copycat brand were present, on average 14% of the time (i.e. they said the brand was present when it wasn’t!).   This error rate increased when the lead brand was present to 25%, regardless of whether the copycat brand was present or not, probably reflecting confusion between the lead brand and other branded products, which is typical within product categories where even major brands share certain design elements. When just the copycat brand was present this error rate rose to 40%, showing that when the lead brand is not available to compare with there is a dramatic increase in the likelihood of mistaking the copycat for the lead brand. But let’s jump back to the 25% error rate when both the copycat and the lead brand were present. From the decision alone we can’t determine whether the 75% of participants who got it right were basing their decision on the lead brand or the copycat brand, for this we need to turn to the eye-tracking data and it reveals an interesting result. When just the lead brand was present, 84% of those correctly identifying it as being present were looking directly at the lead brand when they pushed the button. When the lead and copycat were both present only 68% of those correctly identifying the lead brand as being present were looking at it when they pushed button – 25% of them were actually looking at the copycat brand when they made that decision. In other words, even when given the ability to compare products, even with time to compare products and even with repeated encounters with all the products on the shelves, a quarter of those who said they could see the lead brand, were looking at the copycat brand when they said it. And just in case you think this might be a timing thing, and they had done the comparison by the time they’d pushed the button, when just the copycat brand was present, 73% of those saying the lead brand was present were looking directly at the copycat when they made the decision. What’s more, these errors effects increase further when the images are blurred, confirming that having used the colour to perform the initial search, it’s the details on the pack that are important for accurate decision making.

Now, I’m a fairly moderate scientist, and not given to hyperbole, but this for me is clear evidence that copycat brands disrupt both the search and decision making processes. Moreover, these results would, perhaps unsurprisingly, suggest that those with visual impairment might be more likely to make erroneous purchases. This is bad news for the brands who spend millions on designing and testing their product packaging only to have something remarkably similar, but cheaper, appear on the shelves alongside or, worse still, in place of their products. But, and it’s a big but, this is even worse news for shoppers, because it means they are being played! We know from a considerable, and growing, body of neuroscience research that decision making is mostly done quickly and is highly subject to biases from the ease of collecting the minimum information required to make a correct decision most of the time, the upshot of which is that we sometimes make glaringly obvious errors like those reported here. Now whilst any savvy shopper knows they’re being played by retailers in a variety of ways, a more sinister side to this story is the potential for exploitation of disadvantaged shoppers given that there are an increasing number of products out there whose customers are more likely to suffer visual impairments for the very reason they are in the target demographic. In the healthcare world this is particularly worrying where copycat versions of products already making somewhat dubious claims about benefits for elderly could really cash in!

So, what next? First of all, I challenge you to try to spot these copycat brands when you’re out and about. According a Which? Report last year, there were more than 150 examples in UK stores at the time of publication. I have to confess, ever since I wrote that first blog piece on this topic, spotting them has becomes something of an obsession for me, and a round of “copycat brand bingo” certainly livens up the weekly visit to Aldi for me! Secondly, our study revealed clear steps that could be taken in both design and testing to defend brands against such attacks. Using a range of automated tools and creative market research methodologies, the resilience of a product launch to future attack from parasites can easily be established and I would encourage brand managers to start to adopt such techniques given the perpetual rise of the discounters. Finally, and it should go without saying, take care of your vision! Get your eyes and health checked regularly, and don’t go shopping without your specs, otherwise there might be some unintended surprises when unpacking those shopping bags when you get home.

To those of you in the eye-movement community, I hope to see you all in Vienna next week and watch-out for frequent updates from ECEM on Twitter @DrTimHolmes and blog post or two here!  Oh, and if my mum’s reading this – yes, I’ve started my Christmas list!


Leave a Reply