jump to navigation

N170 face controversy continues June 27, 2007

Posted by Johan in Face Perception, Neuroscience, Sensation and Perception, Social Neuroscience.
1 comment so far

A while back I blogged a paper by Thierry et al (2007 – see also the Neurocritic’s post). Some controversy is brewing about the paper now, so I thought I’d offer an update.

To recap: the Thierry et al (2007a) paper is interesting because it challenges the notion that a specific component of the EEG waveform called the N170 (since it is negative, and occurs at 170 ms) is specific to faces. Thierry et al found that the N170 disappeared when they controlled for inter-stimulus perceptual variance (ISPV), that is, the fact that faces tend to be presented in portraits, while other stimuli are often shown at various angles and sizes.

As the comments on the Neurocritic’s post suggest, some investigators were not entirely convinced by Thierry et al’s (2007a) demonstration. Now one of the critical comments has made its way into the latest issue of Nature Neuroscience, along with a reply from Thierry et al (2007b).

Bentin et al (2007) point to previous research that shows how controlling for ISPV does not in fact explain the N170 specificity to faces. They’ve packed a bit too much information into this figure than necessary, but it’s worth a look:

The black line indicates the mean pixel-wise correlations within each stimulus group, that is, which is an indicator of how much ISPV there was. The grey bars show the N170 amplitude. If Thierry et al (2007) is correct in their argument that the N170 reflects ISPV rather than face specificity, you would expect the black line (indicating ISPV) and the grey bars (indicating N170 amplitude) to match up reasonably well. This isn’t really happening. To make this point even clearer, they show in the supplements that Thierry et al’s (2007) own data seems to indicate the same thing. Bentin et al (2007) also take Thierry et al (2007a) to task for failing to note how their conceptualisation of the N170 contradicts a bulk of well-known effects in previous literature.

So how could Thierry et al (2007a) get such contradictory results? Bentin et al (2007) suggest that one reason may be Thierry et al’s (2007) choice of EEG recording sites, which differ from those generally used by other investigators.

In their reply, Thierry et al (2007b) argue that, among other things, the recording site explanation doesn’t hold, since the same pattern of results appeared across all electrodes. They also contend Bentin et al’s (2007) notion that there is a generally agreed standard for electrode selection.

As for the lack of similarity between the pixel-wise correlations and the N170, this is explained (if you can call it that) by arguing that pixel-wise correlations may not be a perfect measure of ISPV – individual pixels may have very different effects depending on their location, which is not taken into account with such an analysis.

Confused yet? I don’t think there are any clear answers at this point. At the most basic level, the findings of Thierry et al (2007a) contradict previous findings, something that they appear to have failed to mention themselves. There is also some disagreement over what type of analysis is appropriate for this kind of research. Personally, I would like to see an independent replication of the results before I make any further attempts to understand what’s going on.

Bentin, S., Taylor, M.J., Rousselet, G.A., Itier, R.J., Caldara, R., Schyns, P.G., Jacques, C., and Rossion, B. (2007). Controlling interstimulus perceptual variance does not abolish N170 face sensitivity. Nature Neuroscience, 10, 801-802.

Thierry, G., Martin, C.D., Downing, P., & Pegna, A.J. (2007a). Controlling for Interstimulus Perceptual Variance Abolishes N170 Face Selectivity. Nature Neuroscience, 10, 505-511.

Thierry, G., Martin, C.D., Downing, P., & Pegna, A.J. (2007b). Is the N170 sensitive to the human face or to several intertwined perceptual and conceptual factors? Nature Neuroscience, 10, 802-803.


Big brother knows best? Maybe not June 24, 2007

Posted by Johan in Behavioural Genetics, Developmental Psychology, Rants.
add a comment

As a big brother, it is tempting to accept the conclusions that Kristensen and Bjerkedal (2007) draw in their recent article in Science. According to these researchers, IQ is associated with birth order, more specifically social birth order. This measure was created by looking at families where the oldest sibling had died, thus leaving what was biologically the middle child as the “social big brother” (note that the IQ data comes from army conscripts, so all the tested siblings were male). Kristensen and Bjerkedal found that even in these families, the older surviving sibling tended to have a slightly higher IQ than the younger sibling, as the figure below shows.

Note that differing ages is not a factor here, since all siblings were tested at the same age. Kristensen and Bjerkedal argue that this supports a social interpretation of the IQ difference in terms of the family environment, rather than a biological account based on the notion that the first-born might have experienced a better pre-natal environment.

This story has made the rounds both in the news and in blogs, and generally there is surprisingly little criticism. I think Kristensen and Bjerkedal fail to consider an alternative explanation for their results.

Consider the factors that go into deciding whether you want to have another child or not. It is likely that you will consider your experiences with the child or children you already have. Parents who are not as satisfied with their current child or children are less likely to have another child, and this is likely to work the same way when deciding to have child number two, three, and on.

Note that I’m assuming here that parents will pick up on a child’s IQ and that this trait will express itself relatively early on, before the parents decide whether to have another child. So the data for the non-smart first-borns aren’t represented accurately in this analysis, since their parents didn’t have more kids as often.

But here’s the catch: each time you procreate, your chances of hitting the jackpot (ie, all Smart Kids) decreases. This follows from basic probability: if the chance of having one Smart Kid with your genes is x, the chance of having two Smart Kids (x²) must be smaller, and the chance of all Smart Kids continues to decline in this fashion. You’re playing with the same chromosomes each time, so it’s reasonable to assume that the probability is constant.

So if the parents consider their luck before deciding to have another kid, and if they count their luck in the number of Smart Kids, you would expect IQ to drop off as it does in the figure above. Parents who have a first Smart Kid are more likely to have a second child, and parents who have two Smart Kids are more likely to have a third. But with each new child, the chance of the jackpot (all Smart Kids) declines.

To summarise: The parents’ decision to have more children is determined in part by the IQ of the existing children, which means that more intelligent children are also more likely to have younger siblings. But conversely, this selection won’t operate on the youngest child, and will operate to a lesser extent on middle children.

With this account of the data, there is nothing particularly surprising about Kristensen and Bjerkedal’s essential finding that the social big brother (ie, the middle sibling) is smarter than the third sibling. You would expect this, given the interaction between the genetic lottery and the parental choice to procreate.

Let me refute one criticism that could be raised against this account: You might think that the younger siblings in families where the first-born died should have about the same IQ as the younger siblings in families where the first-born didn’t die. This is not the case, as a comparison between the black and the green dots in the figure above shows.

Before you conclude anything from that, look in the supplements for the article. You will find that Kristensen and Bjerkedal restricted their dead first-born sample to cases where the first-born was stillborn, or had died before the age of 1. So maybe the first-born simply wasn’t alive for long enough for the parents to base their further procreation decisions on the smartness of this kid. If this is true, the same parental decision-making process that would normally be based on the first-born is now based on the second-born: Smart Kids are more likely to get younger siblings, while not so Smart Kids are less likely to have younger siblings.

It’s worth emphasising that these are subtle effects. The difference in this study was around 3 points (M=100), so while my discussion of Smart Kids and not so Smart Kids above may sound categorical, I’m only trying to make my point salient. Even if my little theory above is correct, IQ is likely to only play a small role in whether parents choose to procreate or not – otherwise, we would see larger effect sizes in studies such as this one.

If nothing else, I think this study highlights the issue of effect sizes in psychology. Is an IQ difference of 3 points worth discussing? What is the relevance of such a small effect? Can it form the basis of policy changes, or advice to parents? Surely not. I’m not even sure if it has theoretical relevance – surely there are factors out there that explain a bigger part of the IQ variability, and where the exact underlying causes are equally unknown. Yet, this study is treated as if it is hugely important. Look at the quotes below from the New York Times, for instance:

Three points on an I.Q. test, experts said, amount to a slight edge that could be meaningful for someone teetering between an A and a B, for instance, or even possibly between admission to an elite liberal-arts college and the state university, some experts said. They said the results are likely to prompt more intensive study into the family dynamics behind such differences.

“I consider this study the most important publication to come out in this field in 70 years; it’s a dream come true,” said Frank J. Sulloway, a psychologist at the Institute of Personality and Social Research at the University of California in Berkeley.

The edge between liberal-arts college or state university? The most important study in the field for the past 70 years? Don’t believe the hype.

Kristensen, P., Bjerkedal, T. (2007). Explaining the Relation Between Birth Order and Intelligence. Science, 316, 1717.

Let’s play internet memes June 22, 2007

Posted by Johan in Links, Off Topic.
add a comment

Orli at Neurontic tagged me with one of those irritating, yet strangely appealing memes. I’m doing this one out of nostalgia – I spent much of my teenage years filling out inane questionnaires (e.g., do you eat the stems of broccoli?), and this is really more of the same. At least Orli doesn’t threaten 7 years of bad luck if I don’t pass it on…

Superfluous Information About the Author:

1) I do in fact eat the stems. Who doesn’t?

2) There is a defunct 16-channel mixing board in my closet. This is a remnant from a past life, in which I thought I’d be a rock star.

3) I honestly believe that orange marmalade and cheese is the perfect sandwich combination.

4) As a child, my dream job was to be a parking attendant. To this day, my family mocks me about this.

5) I once had a job that involved literally standing by the side of the road, counting the cars that went by. Incidentally, this was shortly before I decided that University would be a good idea after all.

6) My biggest concern at the moment is the health of my succulents. Arty picture below:

I have patiently cared for them since they were just a few centimetres each. I’m soon leaving York for the summer, and they will have to survive for a few months with no attention. I know they’re meant to be desert plants, but I’m not entirely sure that they’ll make it.

7) Like many psychologists keen on neuroscience (I imagine), I have brief moments of despair where I wish I had done a degree in biology instead. Then I glance over to my partner’s textbooks (right now: “The Physiology of Flowering Plants”), and I’m reminded I made the right choice after all.

8) I love cooking. Currently, I’m trying to master the risotto.
I tag:

Bohemian Scientist
Cognitive Daily
Madam Fathom
Memoirs of a Postgrad
Mixing Memory
Peripersonal Space

The Rules:

1. Players start with 8 random facts about themselves.
2. Those who are tagged should post these rules and their 8 random facts.
3. Players should tag 8 other people and notify them they have been

My exams will be over on Wednesday. There might just be some serious blogging after that.

Encephalon #25 arrives June 20, 2007

Posted by Johan in Developmental Psychology, Links, Social Neuroscience.
add a comment

Grab it now at Psyblog. Some favourite posts:

Developing Intelligence reports on some evidence that children have a difficult time telling fantasy from reality. This notion may seem common sense, but this is one of the first empirical demonstrations I’ve heard of.

Omnibrain posted a video of a laughing rat, which almost made me run out and get one for myself.

Finally, Memoirs of a Postgrad explains what embodiment means, as applied in embodied cognition but also in AI research. With mirror neurons being all the rage these days, embodied theories are everywhere. This post is a nice introduction to what the fuss is all about. It’s worth emphasising that embodiment theories preceded  the discovery of mirror neurons, and indeed, it’s not clear that mirror neurons are necessary for embodied representations on a neural level.

Google images now detects faces (poorly) June 17, 2007

Posted by Johan in AI, Face Perception, Off Topic, Sensation and Perception.

Google has somewhat secretly rolled out a special filter for its image searches, that enables you to restrict your searches to faces. Geeks in the know say that this technology comes from their purchase of Neven Vision, a company I’ve never heard of before.

To use this technology, simply add &imgtype=face to the search URL – adding it in the search box won’t work (although Quicksilver users will find that it does work if you use the Google search plugin – I have no idea why).

Face detection algorithms are all the rage in computer science, and there are lots of implementations out there. I thought I’d compare Google to the competition.

An initial test suggests that Google does rather well. Compare a search for “house” with the image filter and without it. There are only faces (including a certain british-comedian turned american-doctor) in the former, and only houses in the latter.

However, it turns out that it’s quite easy to trip Google up once you throw faces of non-human animals at it. According to Google, this is a face:

But it’s not. So how about the competition? There are many online demos of face detection algorithms, but I quite like PittPatt by the Face Group at Carnegie Mellon. You can just plug in a URL rather than having to upload anything, and it’s quite fast. Pittpatt puts rectangles over each part of the image where it thinks a face is. Nothing lights up for the poor dog above.

Let’s make things more difficult for Pittpatt. Gorillas look more like people, right? Google produces mostly faces, but also this:

Ah, Pittpatt seems confused:

Apparently there are a few facial features in there, but I think the blue colour of the frame indicates uncertainty – from the gallery of uploaded pictures on Pittpatt, it seems the most obvious faces are framed in green or at times yellow.

How about chimps, then? Google fails this one spectacularly – all the faces on the first page are non-human, except for a bearded guy and a few pictures of George W. Bush. Go figure. To be fair, this probably has less to do with Dubya’s chimp-like features and more to do with Google’s page rank system, where a picture that is linked a number of times with the word “chimp” nearby ends up on page one of the results for that term. So Bush doesn’t necessarily look like a chimp, it’s just that the Internet thinks he does.

Google thinks this is a face. And you might be inclined to agree, but for the sake of argument I’m assuming that Google wants their algorithm to pick out human faces, not general primate or mammal faces. Pittpatt produces only another weak blue cube, this time restricted to the mouth of the chimp.

So what can we conclude here? This test is of course terribly unfair – I’ve only picked faces that Google failed so at best, Google could only have done as badly as Pittpatt. But it seems like Google’s face detection algorithm isn’t all that great compared to the alternatives. Another possibility is that they’re just trying not to be specist, but I somewhat doubt that. Also, we have learned that a fake statue of King Kong apparently looks more human than a fat dog or a young chimp, according to Pittpatt. I’m not sure what to do with that information.

In any case, it is pretty fascinating that computers can get this good at face detection. While primate faces can look like a human face to us, there is no way you would confuse the two. I guess you might interpret the ambiguous output of Pittpatt as something similar – there is something face-like there, but it’s not quite it.

Update: Other geeks inform me that the face detection algorithm that Google uses is probably trained by either of their two online games Peekaboom or Image Labeler. I couldn’t figure out how Peekaboom works without registering, but the point of the image labeler game is basically for both players to give the same name to a picture. With a training procedure like that, you suddenly start to see how the chimp above might appear under a “face” search – while “chimp” might be the first word you try there, then possibly “ape,” “face” isn’t going to be far behind. This actually points to a problem of human-trained computers – they acquire all our idiosyncrasies and imperfections, for better and for worse.