More Evidence for Separate Systems

In my community of like-minded scientists, we believe that human emotion and motivation is best characterized by two separate systems, one appetitive and one aversive.

We have accumulated a good body of evidence supporting two systems. Particularly interesting are the cases where you engage one system and then call for a response from the other system.
For example, in one study designed by my advisor, Annie Lang, we showed people emotional photographs. During these emotional photographs, we sometimes presented audio tones. Participants were told that they should push a button when they hear the tone.

We stress that the tones are the secondary task. Viewing the pictures is stressed as the primary task. Lang theorizes that some people are more likely to be oriented toward appetitive action, whereas some people are less inclined.

When looking at how quickly these people press the button (a secondary task reaction time, or STRT), those with a tendency toward the appetitive were across the board faster at pushing the button. In writing up these results, we theorize that the button push is an appetitive task, and therefore people predisposed toward appetitive action should be faster.

In a current study in my lab, we tested audio recognition for emotional television material. Participants heard audio snippets from TV they saw and from TV they did not see. They had to say whether that clip was from the TV they saw.

I will save the overall results for another day, but I want to point out the results of the response latencies. Here I am talking about how long it took participants to correctly indicate whether they heard a given clip.

Rather than a simultaneous button push, this recognition task occurred 30 minutes after initial encoding. If our theory holds, people should have been faster at correctly recognizing material from positive content since the recognition cue should -- to a degree -- activate the motivational state at encoding. If that state at encoding was positive, the participant should be shifted toward an appetitive state after the recognition cue. This should make the decision faster since it requires a mouse click, or appetitive action.

This means that our participants should have been faster at correctly recognizing positive content. Further, it would be nice if this difference increased as the clips became more arousing, as extremely arousing clips should induce greater appetitive and aversive activation respectively.



In the chart above, the green line represents correct recognition response latencies from negative television. The blue line represents positive television. On the horizontal axis, the content goes from least arousing (left) to most arousing (right).

Across the board, latencies are fastest for positive content; however, this difference is especially magnified at the greatest level of arousal. Although it would have been nice to have seen this across a gradation, this is nonetheless a nice addition to our ongoing theoretical development.

As a footnote to those who really care, these latency data were cleaned by replacing values that fell outside the range of two times the interquartile range beyond the quartiles. Approximately 3% of values were replaced.