Negative Affect is Stronger than Positive Affect

The following is 6 studies that test whether Negative Affect (NA) is stronger than Positive Affect (PA). These studies examine how we react to negative and positive events in our lives with regards to the intensity of our reactions (study 1), the duration of the reaction (study 2), and our cognitive involvement (experiments 1-4).

Randy Larsen presented these studies in his talk on the Hedonic Treadmill at the 4th European Conference on Positive Psychology.

Study 1 – Reactivity

62 subjects for 56 days were required to keep daily records and ratings of their best and worst events of each day. Therefore, in total 5,971 events were self-rated.

Larsen questioned how good or bad each event was for the average person. He selected the events that were rated as greater than 1 (>1) for standard deviation for his analysis; they were either exceptionally good or bad days.

The following chart shows the results of his analysis:

* These figures are not Larsen’s actual measurements but approximations from a projected powerpoint presentation.

The results of this reactivity study show that there was a significantly stronger reaction for the negative affect self-rating than the positive one.

Study 2 – Duration

In this second study the participants were again required to record how affected they felt by events in their lives. This study lasted 28 days, and required the participants to record 3 times a day their reactions – at noon, 6 pm, and bedtime.

Larsen again selected only the events where the either the PA or the NA had a standard deviation of greater than 1. These events were either particularly negative or positive. Instead of questioning how strongly do people react, he questioned how long does that reaction last?

* These figures are not Larsen’s actual measurements but approximations from a projected powerpoint presentation.

These results show that adaptation rates are far slower for negative than positive events. This idea is what Larsen calls the hedonic treadmill.

In his presentation, Larsen next said that we needed more information to understand the positive and negative affects concerning cognitive involvement. He cites 4 experiments performed by Peter Lang’s Lab in Florida.

Experiment 1 – Emotional priming with sound

Lang used the international affective sound set. (He also has an international affective vision set.) The sound set is comprised of several positive sounds, such as clapping, laughing, and the buzz of a lively room from a cocktail party, and several negative sounds, such as a domestic fight, a car crash, and a baby crying.

In the study, the participants would hear a sound for 10 to 17 seconds and then see a word on a screen, which they would have to characterize as positive or negative within a half-second. Lang analyzed all the errors that the participants made. Here is a graph of the results:

* These figures are not Lang’s actual measurements but approximations from a projected powerpoint presentation.

The results show that the errors, which people made after hearing a positive sound are fairly balanced between the two types of words. However, after hearing a negative sound people made many more errors concerning positive words, i.e. they identified the positive words as negative, and many less errors concerning the negative words, i.e. the correctly identified the negative words as negative.

Therefore, he concluded that the positive system in us seems ready to accept negative sounds.

Experiment 2 – Affective Simon Task

In this experiment participants viewed a variety of negative and positive words, such as murder and happy. They were instructed to ignore the meaning of the word and just to notice whether the word was in upper or lower case. If the word was in upper case, they were to identify the word as “negative,” whereas if it was in lower case, they were to identify the word as “positive.”

He suggested that we cannot help but read and process the meaning of the words. He analyzed the reaction time of the participants. Here is the graph of the results.

* These figures are not Lang’s actual measurements but approximations from a projected powerpoint presentation.

In this graph, the vertical y-axis is the reaction time of the responses. For negative words when they were to respond in “positive,” people’s reaction time was significantly slower. The reaction time was balanced if the words were positive, showing an asymmetry between the positive and negative affects.

We have a senor is our cognitive system moderating the sensory stream, which always asks: “Is this bad for me?” and does not ask: “Is this good for me?” We have more stimuli for processing negative than positive input.

Experiment 3 – Visual Search Task

In the experiment Lang again studied the reaction time of the participants. He set up the faces you see below in a 3×3 grid. They were all the same, or just one face was different. The subjects needed to decide if there was one unique face or not. Half were all the same (same expression, same length of each line, same size of circles) and half were unique in some varying way.

Here are the results:

* These figures are not Lang’s actual measurements but approximations from a projected powerpoint presentation.

The results again show that the reaction time concerning negative stimuli is significantly quicker than the reaction time for positive stimuli.

Experiment 4 – Digital Flanker Task

The participants of this experiment saw many common words with a number at the beginning and end of them. The participants were asked to identify if the number was the same or different. Lang analyzed their reaction time. His hypothesis was that the negative words should decrease their reaction time.

Here are some examples of the words and numbers:

63rotten63
16mama61
32kitten23
83scum88
45lice47
98spa98
91cozy91
63pus63

The summary of the results is below. The y-axis represents reaction time.

* These figures are not Lang’s actual measurements but approximations from a projected powerpoint presentation.

The results show that his hypothesis was correct – reaction time is slower for negative words.