Flashcards in Biological Psychology Studies Deck (58):
What are the studies researching stress?
-Hans Selye (1930, 1950) (GAS model)
-Kiecolt-Glaser et al (1984)
-Cohen, Tyrrell and Smith (1993)
Hans Selye (1930): Aims
To investigate the changes in hormone levels during the oestrus cycle in rats.
Hans Selye (1930): Procedures
Gave experimental rats daily injections of oestrogen and progesterone.
-The control animals received daily injections of saline, (salt water solution).
Hans Selye (1930): Findings
After a few months he found that rats he injected showed a typical suite of symptoms, independent of what they were injected with.
-The rats had stomach and small intestine ulcers and an enlarged adrenal cortex in their adrenal glands. They also had shrunken thymus glands (immune system).
Hans Selye (1930): Conclusions
As both control and experimental animals had the same symptoms it could not be due to the hormones. He concluded that the stress of the daily injection had caused the symptoms. He thought the response was a none specific one, which he called the General Adaptation syndrome, and proposed that humans and animals all responded in the same way.
-In further research, he concluded stressful activities such as extreme cold and fatigue also lead to the same symptoms.
Hans Selye (1930): Criticisms
-Work was done on non-human animals; how far is it reasonable to generalise responses seen to humans?
-Selye failed to consider the role played by the psychological factors in the production of the stress response (anxiety, depression, hostility, threatening, embarrassment etc).
-Selye assumed that all stressors engendered the same response, yet it has been shown that some stressors produce a different pattern of physiological responses.
Kiecolt-Glaser et al (1984): Aims
Aimed to establish a link between stress and reduced immune functioning, based on the assumption that he body's response to stress decreases immune functioning (immune-suppression). They also aimed to establish a difference in immune response between conditions of high and low stress.
Kiecolt-Glaser et al (1984): Participants
A total of 75 first-year medical students (49 males, 26 females) volunteered (a self-selected sample).
Kiecolt-Glaser et al (1984): Procedures
The 75 participants were all under stress due to the imminence of their medical exams, meaning this was a natural experiment because the independent variable was naturally occurring.
-A repeated measures design was used, with participants' blood sample being taken one month before their final exams (low-stress condition) and again on the first day of their final exams (high-stress condition). The numbers of natural killer cells and T cells were measured as indications of immune functioning (the dependent variable), with a high number indicating better immune functioning.
-On both occasions the students were given questionnaires measuring psychiatric symptoms, loneliness and life events.
Kiecolt-Glaser et al (1984): Findings
Natural killer and T cell activity declined between low-stress and high-stress conditions. Thus, the findings confirm the assumption that stress is associated with reduced immune functioning. These findings from the questionnaires revealed that immune responses were particularly suppressed in participants who reported that they were experiencing psychiatric symptoms, loneliness or stressful life events.
Kiecolt-Glaser et al (1984): Conclusions
The research shows that stress is associated with immune-suppression and that the effect is stronger when there are multiple sources of stress.
-A number of different sources of stress were shown to contribute to reduced immune functioning: exams, psychiatric symptoms, loneliness and life events were all implicated. However, only associations were established.
Kiecolt-Glaser et al (1984): Criticisms
-As this is a natural experiment, it means the independent variable was not under the control of the experimenter. As a result, cause and effect cannot be inferred as the independent variable is neither controlled nor isolated (causation can only be inferred when the IV has been directly manipulated). Thus, it cannot be inferred that stress causes the immunosuppression.
-It is important to note that the functioning of the immune system of nearly all of the students was still within the normal range, even in the higher stress condition.
-The immune system is very complex, and so it is hard to be sure that the functioning of the immune system was actually impaired in the higher stress condition.
Cohen, Tyrrell and Smith (1993): Aims
Are stressed individuals more likely to catch a cold than non-stressed individuals?
-The basic strategy was to expose a group of volunteers to cold viruses, and then examine the extent to which they are infected or go on to become ill with a cold.
Cohen, Tyrrell and Smith (1993): Participants
394 volunteers took part at the Medical Research Council's Common Cold Unit, in Salisbury. They had to provide information about three things;
-Number of negative life events they had experienced in the last 12 months.
-Perceived stress, a questionnaire measure of how unpredictable, uncontrollable and overloading the individuals found their lives
-Negative emotions; ratings of the extent to which they had felt 15 emotions over the last week, including distressed, nervous, sad, angry, etc.
Cohen, Tyrrell and Smith (1993): Procedures
The volunteers were then exposed to common cold viruses in nasal drops. The experimental group had drops containing the virus, but the control group had placebo drops.
-This was a double-blind experiment, meaning the cold administrator and the researchers both didn't know whether giving real or placebo, so it wasn't biased.
Cohen, Tyrrell and Smith (1993): Outcomes they were measuring
-Infection; detection of the virus or a significant rise in the levels of virus specific antibodies in the nasal sample, 2-6 days after exposure.
-Clinical colds; a doctor's judgement of cold severity based on symptoms, body temperature, and number of tissues used per day.
-Rates of infection and clinical colds were compared for individuals who scored high and low on the measures of life events, stress and negative emotion.
Cohen, Tyrrell and Smith (1993): Findings
-High levels of stress were more likely to get infection; 90% had infection
-High levels of stress were also more likely to develop a cold; 55% had cold, mostly when bad life events occurred.
-People who had endured chronic stress over a three-month period were three times more likely to develop the common cold than those who had not.
Cohen, Tyrrell and Smith (1993): Conclusions
-There was a clear relationship between the measures and susceptibility to colds, especially when life events, perceived stress and negative emotions were combined in a single scale.
-Perceived stress and negative emotions were associated with infection. Stressful life events were associated with the development of clinical colds, given infection.
-More chronic, enduring stressful events like ongoing marital conflicts and ongoing problems at work put people at risk. However, acute events that last less than a month, such as short term conflicts with a partner, do not have much effect.
Cohen, Tyrrell and Smith (1993): Criticisms
+Cohen was able to establish the kinds of stress that have the most impact.
-We can show there is a relationship between stressful events and the incidence of infectious illness. Stress produces a range of physical and behavioural changes, but those links need not necessarily be mediated by the immune system, meaning causation cannot be found.
-There are some confounding variables, infection rates may relate to the ability of mucosal tissues to block the virus from entering the system or to the production of natural killer cells that destroy viral-infected cells.
-People who are more stressed are statistically more likely to smoke, drink alcohol and have a poorer diet, but no explanation could be found for that.
What are the studies researching stress in everyday life?
-Holmes and Rahe (1967)
-Rahe et al (1970)
-Kanner et al (1981)
-Delongis et al (1982/1988)
-Johansson et al (1978)
-Marmot et al (1997)
-Friedman and Rosenman (1974)
-Kobasa (1979)/Chicago Stress Project (1982)
Holmes and Rahe (1967): Aims
The main aim was to find a way of measuring stress, defined as the amount of change an individual has to deal with during a particular time.
Holmes and Rahe (1967): Procedure
-Holmes and Rahe examined the records of 5000 patients, making a list of 43 life events of varying seriousness that seemed to cluster in the months before the onset of a patient's illness.
-They selected 100 people to help to develop a method of comparing the relative stress engendered by the events. They were told that marriage had been given an arbitrary value of 500. They were asked to assign a number to each of the other life events in terms of 'the intensity and length of time necessary to accommodate... regardless of the desirability of the event relative to marriage'.
-Analysing these judges decisions it was considered that for example, the death of a spouse required twice as much readjustment as marriage.
-From these decisions a weighting scale was constructed by which an individual's stress can be measured in life change units (LCUs), forming a Social Readjustment Rating Scale (SRRS).
Holmes and Rahe (1967): Findings
-When applied to their patients, Holmes and Rahe found evidence to support their belief in the relationship between stress experienced and illness. There was a small but significant correlation between score of the SRRS and later health problems.
-Individuals scoring 200-300 in the previous year were statistically likely to develop health problems in the following year, it indicated a 50% chance of developing a health problem.
-People with a score of over 3000 increased this probability to about 80%.
Holmes and Rahe (1967): Conclusions
Holmes and Rahe concluded that stress could be measured objectively as an LCU score, that is a number corresponding to the amount of change a person has to face and adjust to during a 12 month period. This in turn predicts a person's chance of becoming ill (physically and mentally)
-Stress and illness appear to be correlated but stress appears to be responsible for the illness, although this is not always the case.
Holmes and Rahe (1967): Criticisms
+Provided a way of measuring stress die to life events (LCUs), modified scales for different groups shows same effect, used in different contexts, e.g. students.
+A link was found between life events and subsequent ill health.
-Findings are correlational so cause and effect can't be established.
-Ratings for individual events are arbitrary and may vary from person to person as people are affected differently by the same events, e.g. divorce.
-Started with participants who were already ill, meaning it was biased, but did lead to Rahe et al.
-No distinguished differences between +ve and -ve events, negative events are now considered more significant.
-Original SRRS scale not applicable to all populations (can't generalise and only applicable to white professional males)
-Individual differences, same events can have a different impact on different people.
Rahe et al (1970): Aim
To find the association between stressful life events and illness
Rahe et al (1970): Participants
2500 male US naval personnel over a period of six month (longitudinal)
Rahe et al (1970): Procedures
The number of life events of the participants was assessed using a self-report questionnaire based on the SRRS. Each of the 43 life events had assigned a value of LCUs, based on how much readjustment the event would require. A total life change score was calculated for each participant adding up the LCUs of each life event. A health record was also kept of each participant during the six-month tour of duty.
Rahe et al (1970): Findings
A correlational analysis was carried out to test the association between total LCUs and incidence of illness. A significant but small positive correlation of +0.118 between the total LCU score and illness. As total score increased, so did illness.
Rahe et al (1970): Conclusions
The positive correlation between total LCUs and incidence of illness was small but significant due to the large amount of participants.
Rahe et al (1970): Criticisms
+Important due to the association between stress and illness.
+Improvement on previous research (Holmes and Rahe) as it was over a longer period.
+Cause and effect more or less established, physical health assessed after the life event, increases chance life events were helping to cause problems.
-Cannot completely establish cause and effect, IV not manipulated directly/only correlational data
-Illness is likely to help cause certain life events, e.g. change in eating habits
-Biased, only American men used, Ethnocentric (one culture) and androcentric (only males).
DeLongis et al (1982): Aims
Thought relative rarity of life events explained the weak correlation between life events and health. Investigated whether or not daily hassles would be better predictors of later health than life events.
DeLongis et al (1982): Procedures
For a period of a year, their participants were required to complete monthly measures of hassles and uplifts, life events and health status.
DeLongis et al (1982): Findings
-They found a significant positive correlation between intensity of hassles and health status.
-No relationship was found between health and uplifts.
-No relationship between life events and health, during the period of study.
DeLongis et al (1982): Conclusions
The daily hassles and uplifts scale may be a better way ti predict health consequences of stress than the usual life events approach.
Johansson et al (1978): Aim
To investigate the workplace stressors in a Swedish sawmill
Johansson et al (1978): Participants
14 finishers in a Swedish sawmill, whose work was very stressful, high repetitive and machine-paced. The rate at which these individuals completed their part of the job determined the overall productivity of the sawmill, and therefore everyone else's wages.
-They were compared with a group of 10 low stress workers (cleaners)/
Johansson et al (1978): Procedure
Johansson et al measured levels of stress hormones (adrenaline and noradrenaline) during work days and rest days. They also looked at other measures such as sickness and absenteeism.
Johansson et al (1978): Findings
-The finishers had raised levels of stress hormones on work days (and also on rest days)
-They had a higher rate of stress-related health problems
-There was a higher rate of absenteeism for this group than a control group of workers from other jobs in the sawmill.
Johansson et al (1978): Conclusion
The researchers concluded that the work environment of these workers was particularly stressful because:
-They were responsible for the wages of the whole factory.
-Their work was boring and repetitive.
-The work was machine-paced, therefore they lacked control over their work environment.
-They worked in social isolation.
They also concluded that finding a new way of wages (set weekly rate), improve productivity, job rotation and by giving them control over production line would help decrease stress levels.
Johansson et al (1978): Criticisms
+High ecological validity as it is an experiment in a real work situation, meaning it can be generalised to other work contexts.
+Measure of stress hormones in the urine is an objective measure of stress, It reduces the change of investigator effects and has higher validity than self report measures of stress levels.
-There no way to conclude the biggest stressor of the workplace, as some stressors may have a bigger effect than other and therefore are more important to reduce.
-Large difference between the experimental and control group. The finishers were highly skilful and knowledgeable whereas the cleaners were unskilled and much more social.
Marmot et al (1997): Aim
To examine the relationship between control in the workplace and risk of developing CHD.
Marmot et al (1997): Participants
7,372 civil servants, aged 35-65, in a longitudinal study lasting 5 years.
Marmot et al (1997): Procedures
Participants carried out a self-report survey (questionnaire), independent assessments of the work environment by personnel managers. They were assessed on two occasions, three years apart.
Marmot et al (1997): Findings
They found a negative correlation between control and illness.
-Participants with the least control had a 1.5 to 2.2x inflated risk of CHD symptoms, compared with those with the highest control.
-Low control workers were four times more likely to die of a heart attack and suffer stress-related illness (cancer, stroke, gastrointestinal disorders) than those with high control.
Marmot et al (1997): Conclusion
They concluded that lack of control over work content and rate led to increased stress, which led to higher risk of illness.
-They also suggested to improve employers could address lack of job control, by giving employees more autonomy and control. This is likely to decrease illness as well as increase efficiency, productivity and general wellbeing.
Marmot et al (1997): Criticisms
-The evidence is correlational which means we can't be sure that low job control causes stress related illness. Those whose health is poor are less likely to achieve career success and have jobs offering good control.
-The sample were all civil service employees so although it has high ecological validity due to being carried out on real workers' jobs, it can't be generalised easily to different workers as they are required to do different skilled/stressful/difficult jobs.
Friedman and Rosenman (1974): Aims
To investigate links between the Type A behaviour pattern and cardiovascular disease.
Friedman and Rosenman (1974): Participants
Self-selected sample of 3200 Californian men aged between 39 and 59 at the onset of the study. This was the age they were most likely to develop and die from cardiovascular disease.
Friedman and Rosenman (1974): Procedure
-This study was a prospective longitudinal study in which the participants (all healthy at the start) were assed over 8 and half years.
-Part one of the study included a structured interview and observation to assess personality type and health status. Personality type was determined by the amount of impatience, competitiveness and hostility reported and observed during the structured interview and from participants' answers to questions. During the interview, the interviewer deliberately interrupted the participants and looked for signs of impatience or irritability.
-On the basis of the structured interview, participants were classified as A1 (Type 1), A2 (Not fully type A), X (equal amounts of Type A and B), and B (fully Type B).
-Part two of the study was the follow up 8 and a half years later, when incidence of coronary heart disease was recorded. A correlational analysis was carried out to test the association between Type A/B personality and coronary.
Friedman and Rosenman (1974): Findings
257 men had developed coronary heart disease during the 8 and a half years of the study. 70% of those who'd developed CHD had been classified as Type A, nearly twice as many as those who had been classified as Type B. This remained the case, even when other factors associated with the disease were taken into account. Compared to Type B's Type A's were found to have higher levels of adrenaline, noradrenaline and cholesterol.
Friedman and Rosenman (1974): Conclusion
Type A personality is fairly strongly linked to coronary heart disease. They concluded that the Type A behaviour pattern increases the individual's experience of stress, which increases physiological reactivity which in turn increases vulnerability to coronary heart disease. The stress response inhibits digestion which leads to the high level of cholesterol of the blood. This places Type As at risk of coronary heart disease.
Friedman and Rosenman (1974): Criticisms
+It has real life implications, as they have identified Type As have this life threatening risk, people are aware and they can now aim to reduce it.
-Type A personality consists of several aspects and it wasn't clear which of them were the most important in producing vulnerability to CHD. As a result, this research lacked internal validity because it didn't measure precisely what it intended.
-Research is correlational in nature so we can't be sure Type A causes coronary heart disease, it is likely there are other alternative factors.
-Assigning everyone to one of four categories (Types) is oversimplified and dubious.
Kobasa described people as being hardy or non-hardy. There are three main characteristics of hardy personalities:
-Commitment: very involved in what they do
-Challenge: see challenge as an opportunity to develop themselves
-Control: Internal locus of control
Maddi et al (1987)
Maddi et al (1987) looked at employees of a US phone company that was reducing its workforce. During the downsize, two-thirds of the workforce suffered from stress-related health problems. However, the other one-third of its workforce thrived during the period. It was found that these people were much more likely to have a hardy personality.
-To support the use of benzodiazepines
Kahn (1986) investigated 250 patients over 8 weeks and found that benzodiazepines were more effective at reducing stress than placebos and other drugs. Such research gives the drug scientific credibility.
Sheehy and Horan (2004)
-To support SIT
Sheehy and Horan (2004) gave first year law students weekly SIT sessions lasting for 90 minutes each. The session aimed to reduce their levels of anxiety and stress in order to increase their academic performance. After 4 weeks it was found that all participants had lower levels of stress and anxiety, and that many students improved their academic performance.