To kill bias, gather good data

I hate myself for this: I have the worst sense of direction.

For the entire year when I was living in my first apartment in Kingston, I would take a circuitous route along King Street and then up Princess on my way home from the Kingston Yacht Club. Nearly two kilometers, when walking up West Street would have got me home in half the time. As Charlie said when I revealed this to him, “Two sides of a triangle is always greater than one.”

It’s not that I didn’t know grade school geometry, or that I wanted a more scenic route. I just stuck to the path I knew would get me there.

I felt a bit triumphant when I realized how long it can take Charlie when you ask him to pick up the milk. The last time I dragged him to the grocery store, I left him alone for a few minutes to use the bathroom, and returned to find him loading pineapple after pineapple after pineapple – painfully slowly, into the cart. We laughed, but I don’t ask him to come with me anymore. Alone, I can collect a week’s worth of food in less than 20 minutes.

I’m not ashamed to admit my navigational failings, either. My field assistant Myra and I happily agreed that our best strategy driving around Los Angeles was that we should always do the opposite of whatever we both thought was correct. It worked.

What I hate is my sneaking suspicion that I’m just a lame stereotype. Maybe I’m a terrible navigator because of biology; female brains are just not suited for getting around.

Hunter, gatherer

Modified from this cartoon. Original source unknown.

Recently, psychologists looked at this sex difference in what seemed like a neat field study of human foraging behaviour – in a grocery store1. Joshua New from Yale University, and his coauthors from UC Santa Barbara, set up a unique experiment in a California farmers’ market: they led men and women around the market, giving them samples like apples, fennel, almonds and honey. Then they brought the subjects back to a central location and asked them to point in the direction of those same food items.

These researchers wanted to test the idea that women outperform men at certain kinds of spatial tasks: while men are thought to be better at vector-based navigation, women might excel at remembering the locations of objects, because of the way foraging roles were divided up when our brains were evolving. It’s thought that in our hunter-gatherer past, big game hunting meant that men had to figure out how to bring heavy prey home by the most direct route. Women foraging closer to home needed a much different set of spatial adaptations2. It’s not that men are better at spatial reasoning in general, you just have to choose the right task3.

And in the huge Santa Barbara farmers’ market with close to 100 stalls, women did just that, pointing towards the locations of foods they had sampled with 9 degrees greater accuracy, on average, than men1. On top of this, everyone did better at pointing out the locations of energetically rich foods. People remembered almonds, honey and olive oil more accurately than lettuce, cucumber or zucchini.

Joshua New and his coauthors promoted this as the first time anyone has bothered to look for a female navigational advantage at the right scale. Previously, most research on object-location ability used the memory task where you briefly look at a set of items like keys, pencils, and coins on a desk, and then have to figure out which one is missing later on. These studies typically show mixed results when it comes to sex differences4. The farmers’ market seems to offer the advantage of being an ecologically relevant task – field biology of urban humans. The researchers also claimed that, because people were much better at locating high-energy foods in the market, the task must be drawing on our psychological adaptations for efficient foraging.

But if you read the paper, there are some gaping holes in the logic that make me bristle. The biggest one? They fail to consider the alternative that there are plenty of other reasons why women would excel at their grocery shopping task.

Sure, they tried to control for some of the factors that could bias the task towards women. First off, they designed the pointing test to favour the kind of vector-based navigation generally preferred by males. They also asked the men and women to rate their own sense of direction on a scale from 1-7 afterwards, since previous research suggests that people are quite good at assessing their own navigational abilities1. And, consistent with other studies, the men in this group rated themselves much higher than the women. Clearly, the results weren’t an anomaly of catching an unusually direction-savvy group of women in the market that day.

But both of these methods only provide a semblance of scientific control when learning and prior experience could have had a much bigger effect on performance.

As the authors mention in their discussion, women are way more experienced in food markets, since they are the primary food shoppers in about 73% of US households5. But they quickly dismiss this from affecting their results, citing research showing that women are no better than men at navigating other real-world shopping scenarios. None of this surprises me: I’m hopeless in a Canadian Tire, mall or big box store, but I’m pretty efficient in Loblaws. And it’s not due to specific experience, since I do my grocery shopping elsewhere. When they re-arranged the Kingston Loblaws store a couple of years ago, it only took me one or two visits to figure it out. They did the same thing at Canadian Tire and I barely noticed, even though I’m in there about as often.

The point is that in Canadian Tire, I don’t have learning down to routine. Food shopping is different: several times in my life I’ve had to adjust to a new grocery store. I do it pretty quickly by now, and wouldn’t be surprised if I have a mental program for it. My beef with the farmers’ market study is that the authors never consider the obvious alternative: that women, as the more experienced group, have learned something about learning in this scenario. Practice might help build the scaffolding you need for remembering certain kinds of things, like the contents of your fridge.

And besides, a lack of female advantage in other shopping scenarios could just as easily be taken as evidence against the hunter-gatherer hypothesis. If women have better object-location memory for things on the scale of a few hundred metres, shouldn’t it translate to those other shopping scenarios, too? If not, why do the supposedly male-typical spatial abilities work in novel situations, like driving or rotating 3D objects? The authors never consider this.

Instead, they push the idea that their farmers’ market task is just right for looking at foraging adaptations, because participants were so much better at finding the nuts and honey than the vegetables. Perhaps, but it doesn’t follow that the sex differences the task calls up are also necessarily foraging adaptations. Besides, there are alternative explanations for this part of the experiment too, including some obvious psychological confounds. Produce differs from the other rich food items in many ways: colour, texture, absence of packaging, not to mention degree of specialization of the market stands that typically sell these things. How often do you see a market stand that sells just zucchini? Now picture the guy who sells honey. See? All of these cues might cause the location of nuts and honey to be more memorable than the lettuce.

There may also be reasons to doubt the validity of the hunter-gatherer model of spatial cognition as a starting point. Although most current human societies evidently do have a strong division of labour among the sexes, this might be a fairly recent development. A 2006 review of the archaeological evidence suggests that that the traditional hunter-gatherer sex roles did not emerge until the Upper Paleolithic period, around 45,000-10,000 years ago6. Furthermore, Neanderthals, our stone age cousins with whom we shared Europe and Asia for over 100,000 years, showed no such division of labour6. Neanderthal women were robust. The ladies had man-hands, and their remains often show injuries that are the tell-tale signs of big game hunting. Given that ancestral humans have been hunting with tools for well over 2 million years, how important could a few thousand generations of hunter-gatherer sex roles have been for brain evolution in the first place?

So here’s what we’re left with: women, who are generally more experienced at finding groceries, were better at finding groceries. If you already know what you’re going to get, is it science? Unfortunately, this kind of handwaving is typical of a lot of evolutionary psychology, which seeks to explain the mind based on hypotheses about what the environment might have been like for early humans.

But don’t get down on science. There is plenty of good research, too. Briefly, here are some tidbits about what else is going on in this field:

  • Anthropologists are using MRI scans and students on treadmills to look at early human adaptations for running7. Researchers at Arizona State University scanned young cross-country runners in an MRI machine to measure their calcaneus, a hard part of the heel. It happens to fossilize, so they had calcaneus measurements from a number of ancestral hominids, too. They showed that the runners’ oxygen consumption on the treadmill was negatively related to calcaneus length. Shorter ones are more efficient. Early humans, who had short calcaneus lengths and lived in warm climates, may have been adapted for endurance running and pursuit hunting. Neanderthals, with longer ones relative to body size, were not.
  • The size-weight illusion is an error that almost all people make. If you lift two objects of equal mass (say, a pounds of flour and a pound of lead), you’ll most likely perceive the larger one as being lighter. Qin Zhu and Geoffrey Bingham showed that this universal error might be a byproduct of our other cognitive adaptations8. Specifically, mental programs for throwing, since there is a near-perfect correspondence between the mass of your preferred projectile and the way you experience the size-weight illusion. The paper is a terrible read, but Zhu and Bingham tested men and women who were not experienced throwers. And in other experiments, they have looked at the effect of throwing practice on the size-weight illusion (the short answer is that there isn’t one)9.
  • Neanderthals had big noses. For a long time, this was thought to be an adaptation for the cold ice age environment. But until recently, no one bothered to check – problematic since most arctic mammals actually evolve narrower noses in response to cold9. Researchers used CT scans to measure the sinus cavities of Neanderthal and other human skulls in 3D, using fossils from sites in northern Europe that had similar climate history. They showed that Neanderthal sinuses are not any bigger when you adjust for overall skull size. In addition, all of their samples fit into the same general relationship between skull and sinus size. The conclusion? The distinctive Neanderthal face cannot be explained by adaptation to cold after all.

These examples demonstrate that it is possible to use fancy new technologies, and some pretty basic ones, in more rigorous ways to study human origins. The results give a less biased picture of what early hominids might (and might not) have been good for.

Maybe women are biologically predisposed to the gathering role, but that doesn’t mean that all of the sex differences that come along with this division of labour are hard-wired into our brains – and failing to consider the alternatives is no place to start.


  1. New, J. et al. 2007. Proceedings of the Royal Society B 274: 2679-2684.
  2. Murdock, G. P. 1967. Ethnographic Atlas. University of Pittsburgh Press, Pittsburgh.
  3. Silverman, I. and Eals, M. 1992. In: The Adapted Mind (eds. Barkow, J. et al.). Oxford University Press, New York.
  4. Postma, A. et al. 2004. Brain and Cognition 54: 24-34.
  5. International Mass Retail Assocation. 1993. The 1992 consumer research study. In: IMRA 1993 Washington, DC.
  6. Kuhn, S. L. and Stiner, M. C. 2006. Current Anthropology 47: 953-981.
  7. Raichlen, D. A. et al. 2011. Journal of Human Evolution 60: 299-308.
  8. Zhu, Q. and Bingham, G. P. 2011. Evolution and Human Behavior.
  9. Zhu, Q. and Bingham, G. P. 2010. Journal of Experimental Psychology: Human Perception and Performance 36: 862-875.
  10. Rae, T. C. et al. 2011. Journal of Human Evolution 60: 234-239.


  1. I always knew there had to be a good reason why I always drive and Ken always navigates. When we hike, he studies the contour maps and I bring the food.

  2. Good one. You offer an interesting and pleasing perspective on the whole subject of sex differences, and human navigation in this case.

    I wonder how long before I read this article in a magazine?

Comments are closed.