Can We Predict When People Take Risks: The Impact of Individuals' Mood States on Risk Taking. P. J. Clough and G.R.J. Hockey, Department of Psychology, University of Hull, UK; and A.J. Maule, Department of Business and Economic Studies, University of Leeds, UK
Introduction
Risk can be identified as a feature of personal decision making
which depends not only on stable differences but on variability
in emotional states, mood, stress and other conditions. Emotion
is thought to influence risk by changes in the pattern of information
processing, notably in (a) the costs and benefits associated with
different outcomes of actions (Maule & Hockey, 1993) and (b)
attentional resources available for the decision process (Matthews,
1992). The reported research examines the effect of mood on decision
making, using naturally occurring mood states.
Defining Mood
For our purposes mood is conceptualised in terms of two independent
dimensions, negative and positive affect (PA and NA), proposed
by Watson and Tellegen (1985). PA is characterised by a contrast
between terms such as enthusiastic, cheerful, energetic (FUGH)
and depressed, miserable, weary (LOW); NA refers to being anxious,
tense, annoyed (vs calm, relaxed, at ease). PA is sub-divided
into (a) PA1- the standard emotion related measure, associated
with active goal orientation (enthusiastic and depressed) and
(b) PA2- a more abstract energy measure (energetic, lively, alert
vs tired, fatigued, weary).
Measuring Mood
An important methodological feature of this research is the need
to 'calibrate' within-person reports of moods, in order to take account of differences between
individuals in their baseline level and variability of
reported affect. Single individual reports of state measures
are generally unreliable and subject to confounding from large
stable differences, e.g. in reporting bias (Depue & Monroe,
1986). While such differences may be of interest in some instances
they reduce the power of within-person analyses. Alternatively,
observed effects may be artifacts of stable trait differences,
rather than the results of state changes. In our studies baselines
are therefore established by having participants complete mood
diaries for between 2 to 9 weeks. Then, instead of simply using
the raw mood scores, measures of 'relative affective state' within
the person's own experience and reporting style are calculated.
Measuring Risk
One of the methods used to assess risk taking behaviour has been
the Personal Risk Inventory (PRI); a short self report instrument.
This was developed as follows. A corpus of personal risk items
was generated by advertising in local press and radio, telephone
sampling and local contacts. A total of 200 questionnaires, 330
telephone contacts and 40 direct verbal reports were initiated.
Respondents reported everyday situations in which they had to
make a decision that involved both a risky and a safe option.
These were classified in terms of their major themes as belonging
to one of 5 categories of personal risk: physical/health; social/moral;
financial; legal and novelty.
A set of 40 items were generated (2 parallel sets of 20). Each scenario had a choice of two actions (one identified as risky), with level of commitment to the choice and perceived risk also being measured.
Risk and mood in a sample of Industrial Chemists: An illustrative example
A number of major field studies have been undertaken at the University of Hull, funded as part of the ESRC Risk & Human Behaviour' project, examining mood and risk taking using a wide variety of samples (students, the general population, G.P.s and other professionals). To illustrate the approach a summary of an investigation of risk taking in industrial chemists is reported below.
An opportunity was taken to test 2 groups of management level industrial chemists, taking part in a professional development course. Thirty three participants completed the PRI twice, at the be beginning of the 5 day course, and at the end of the second day, after failing a management exercise involving the whole group as a team. Mood diaries were completed at the time of the PRIs and also for a three week period a month or so after the training week (to provide individual calibration baselines). The failure experience induced a complex and severe change in mood states, in the form of reduced PA1/PA2 and increased NA, as well as a marked increase in risk. There were strong correlations between changes in moods and risk. Those with the most marked mood changes also showed the strongest shift towards riskiness (for NA, r=0.44; PA r=-0.42). Risky behaviour was found to be promoted by an increase in both anxiety (NA) and depression (low PA1).
Conclusions: Do moods impact on risk taking behaviour?
As previously stated a number of studies have been undertaken using the PRI and mood diaries. In addition, a series of laboratory based investigations have also been carried out by colleagues at the University of Leeds. It is possible at this time to draw some tentative preliminary conclusions.
Mood states do appear to be associated with risk taking behaviour,
although the relationship is quite complex. When mood states
are classified as joint changes in NA and PA there appears to
be an interaction in their effects on risk. Preliminary analysis
suggests that when people are relaxed (low NA) they are more risky
when they also feel cheerful or energetic (high PA1/PA2); when
they are tense (high NA), however, riskiness is greater when they
are also depressed or tired (low PA1/PA2).
References
Depue RA & Monroe SM (1986) Conceptualisation and measurement
of human disorders in life stress research: the problem of chronic
disturbance. Psychological Bulletin, 99, 35-51.
Matthews G (1992), Mood. In AP Smith & DM Jones (eds.), Handbook
of Human Performance. Vol. 3. Academic Press: London.
Maule AJ & Hockey GRJ (1993) State, stress and time pressure.
In O Svenson & AJ Maule (eds.), Time pressures and stress
in human judgement and decision making. Plenum: New York.
Watson D and Tellegen A (1985) Towards a consensual structure
of mood. Psychological Bulletin, 98, 219-235.