Cognitive Daily

Take a look at these two photographs of my son Jim taken a month or so after he was born (and, as he would be quick to point out, nearly 14 years ago). Which is more memorable?


It may depend on your age. It’s natural for your priorities to change as you get older, and so it seems, you may have a different response to pictures depicting emotions. Your kids grow up and leave home, and suddenly Little League and Disney seem less significant. Perhaps fine wine and opera rise up to fill that void. Later still, you begin to think about retirement, and gradually it seems more important to reach out to family and friends. Laura Carstensen was part of a group of researchers who developed a “socioemotional selectivity theory” to explain these changes—they argued that our emotion-related goals increase in importance as we age because we are assessing how much longer we have to live.

As we age, they claim, emotion increasingly becomes the central motivator for a wide sphere of actions, from who we choose to spend time with to how we deal with problems. Perhaps surprisingly, this reorienting of goals around emotions, all motivated by impending death, leads to the result that older adults are better off emotionally than younger people.

Carstensen joined with Martha Mather to expand on these conclusions—they wanted to see if this focus on emotions in older adults also affected cognition, and so they developed a simple reaction-time and memory test. Two groups of people participated in the study: older adults with an average age of 74, and younger adults averaging age 26. They were questioned about their emotional state, and consistent with socioemotional selectivity theory, the younger adults scored significantly lower than older adults on an index of negative emotion.

The participants were shown pairs of photos of 60 different faces. Each pair of photos depicted the same person: one with a neutral expression, and one showing an emotion— happy, sad, or angry. The photos remained on screen for 1 second, then disappeared. In place of one of the photos, a small grey dot appeared, and participants were asked to press a button on the keyboard to indicate where the dot had appeared (left or right).

As you might expect, the younger adults responded much faster than older adults—in about 420 milliseconds compared to 780 for older adults. But when responses to emotional faces were compared to those for neutral faces, another striking difference appeared:


These results were obtained by subtracting the average reaction time for the emotional face from the reaction time for the neutral face in a pair. For younger people, there was no significant difference in reaction times between emotional faces and neutral faces, or even between positive and negative faces. For older people, the emotion difference in the faces caused a comparatively large difference in reaction times. The slowest reaction times were always for negative faces, and the fastest times were for positive faces.

After participants completed the reaction time test, Mather and Carstensen tested memory. They showed viewers a new set of paired photos, each containing one previously-viewed face and one new face. The emotions in these new pairs of photos always matched: the task was to indicate which face they had seen before. Here are the results:


When the emotion for the test pair was happy and the original pair of photos had been happy/neutral, older adults were significantly more likely to correctly recall the old picture than for any other condition. While younger adults were somewhat less likely to remember old negative faces, for the most part the emotion of the faces did not impact their memory for faces.

Mather and Carstensen conclude that these results show not only that subjective attention to emotional issues change as we age, but also the way our basic cognitive processes work when we are confronted with emotional images. Motivated by impending death, emotion affects our life priorities, but these important changes as we age also impact our ability to react to and remember simple images.

Mather, M., & Carstensen, L.L. (2003). Aging and attentional biases for emotional faces. Psychological Science, 14 (5), 409-415.


  1. […] Take a look at yesterday’s post in Cognitive Daily. Like most psychology research, this study found no racial differences. Like many studies, this one actively tried to find differences, but observed none. It found that as we age, whether we’re black or white, emotion begins to color our memory and reaction to faces. Black or white, our behaviors are the same, as they are in countless other studies. But Pinker chooses to focus on one study that tries to explain why there are so many Jewish lawyers and doctors. Did Pinker suddenly learn that Jews tend to score better on IQ tests than other racial groups? I’d submit that he didn’t he’s just talking about it more now. […]

  2. #2 Terri Joy Holmes
    December 7, 2005

    Perhaps it would be wise to consider the stressful demands placed on a 27 year old functioning within U.S. culture. The daily demands of home, family, children job, debt and a mass of other problems – leaves little time or energy for reflection.

    Too, negative thoughts maybe the result of being taken advantage of – an internal signal warning a person to address the problem. Unfortunately having been taught to obey and comply since infant/childhood, does not allow a person to learn how to ‘confront problems’ when power and control over one’s own life is in the hands of another.

    The needs, desires and expectations of the elderly are so vastly different, there can be no comparison.

  3. #3 Dave Munger
    December 7, 2005


    The study doesn’t really address whether younger adults are justified in having more negative thoughts than older adults. I think you’re right to question the socioemotional selectivity theory, however. Though this study does seem to add support to the theory, and even though the theory is based on the results of several different studies, it’s certainly possible that the differences between older adults and younger adults are so complex that a single theory couldn’t explain them.

    On the other hand, it’s also possible that the results here are due to shared physiological differences between the young and the old—that indeed, it’s not as complex as it seems. Until we find research results that contradict the theory, we must acknowledge that it does have impressive predictive power.

  4. #4 Mara Mather
    May 29, 2006

    Interesting blog! I’ve been enjoying it for a while and just came across this post about my research. In a recent paper (Mather & Knight, 2005), we present evidence that older adults’ positivity effect requires cognitive control and is not the result of cognitive decline. The study is described briefly here:

New comments have been disabled.