(AQA A) AS Psychology

Unit 1: PSYA1 – Cognitive Psychology, Developmental

Psychology and Research Methods

Key Study Title Relates to:

Miller (1956) Capacity of STM

Baddeley (1966) Semantic Encoding in LTM

Conrad (1964) Acoustic Encoding in STM

Peterson & Peterson (1959) Duration of STM

Bahrick (1975) Duration of LTM

Loftus et al (1978) Misleading information

List (1986) Factors affecting accuracy of Eyewitness testimony (EWT)

Poole and Lindsay (2001) Factors affecting accuracy of Eyewitness testimony (EWT) – Age of witness

Fisher et al (1989) The use of the cognitive interview

Key study Title Relates to:

Ainsworth and Bell (1970) Types of attachment

Schaffer and Emerson (1964) Development of attachment

Harlow (1959) Learning theory of attachment

Bowlby (1944) Disruption of attachment

Hodges and Tizard (1989) Failure to form attachment and the effects of

institutionalisation.

Rutter et al (2007) Failure to form attachment and the effects of

institutionalisation.

Takahashi (1990) Cultural variations in attachment

Van Ijzendoorn and Kroonenberg (1988) Cultural variations in attachment

Andersson (1992) Implications of research into attachment and day care

for child care practices.

Belsky and Rovine (1988)

Violata and Russell (1994)

NICHD report (2001)

The negative impact of day care on children’s social

development (including peer relations and

aggression).

Schweinhart et al (1993)

Creps and Vernon-Feagans (1999)

Vandell et al (1988)

The positive impact of day care on children’s social

development (including peer relations and

aggression).

Capacity in STM

Miller (1956)

Why are there so many things that come in sevens – such as seven wonders of the world, seven

deadly sins, seven primary colours, seven notes of the musical scale, and seven days of the week?

Miller (1956) reviewed psychological research to see what had been discovered. Investigations into

various cognitive abilities have found that people can easily distinguish three musical tones, but

when they asked to distinguish between five or more tones they become progressively more

confused. If research participants are shown an arrangement of dots, flashed onto a screen, they

can count five or six of them easily. More than that number and accuracy deteriorates. Such findings

suggest that our span of absolute judgement (e.g. distinguishing musical tones) and of immediate

memory (e.g. counting dots) is for about five or six items. This would lead us to expect that, if you

have to remember a string of letters, seven letters might be the maximum that could be held in

short-term memory. But what if the letters were C-A-T-W-A-L-L? In this case the seven letters make

two chunks of data (“CAT” and “WALL”). All words consist of bits of information – in the case of

“catwall” there are seven bits altogether, but there are only two chunks of information. If memory

was limited by “bits” of information it could handle, then we can predict that it would be possible to

remember twice as many five-letter words as ten-letter ones. Miller found that this wasn’t true.

Research has demonstrated that people can remember about seven words, no matter how many

bits of information are in the word. It is the chunks of data that limit memory. Miller called chunking

“the very lifeblood of the thought processes”. It improves the capacity of memory, although it may

reduce accuracy.

Evaluation of Miller (1956)

This research has useful applications. For example, phone numbers are chunked to make them

easier to remember when making a phone call.

However, later research didn’t completely confirm Miller’s conclusion. For example, Simon (1974)

found that the number of bits in a chunk did have an effect on memory. Participants in his study had

a shorter span for larger chunks such as eight-word phrases, than for smaller chunks such as one-

syllable words. Therefore, immediate memory span is related to chunking but the number of chunks

remembered depends how big the chunks are!

Some psychologists, such as Case (1974), have used

the idea of chunking to explain how children become

capable of more complex mental tasks as they get

older. Case suggests that M-space (which is roughly

equivalent to STM) is used to hold information

required while you are thinking. The restricted size of

this space acts as a limit to cognitive processing. As

one gets older the space increases a little, but more

importantly we also become more efficient at using it.

An example of this would be playing chess.

Inexperienced players have to hold the rules in M-

space while they are playing. As one becomes more

experienced these rules become more automatic,

having been incorporated in higher-order units – a

process akin to chunking. This form of chunking leaves more space in M-space for mental activities.

Task: With a learning partner, find a strength and a weakness of Miller’s findings. (Remember he reviewed others’ work). Does the research into capacity of STM have ecological validity? Can it be applied to real life?

Evaluation of Baddeley (1966)

The use of the experimental method allows a causal

link to be drawn between type of coding used in STM

and LTM and the accuracy of recall.

The conclusions of this study may not reflect the

complexity of encoding. Evidence from other studies

shows that, in certain circumstances, both STM and

LTM can use other forms of coding.

Semantic Encoding in LTM

Baddeley (1966)

The aim of this study was to explore the effects of acoustic and semantic encoding in STM and LTM.

In the STM study, participants were asked, immediately after presentation, to recall, in serial order,

a list of five words taken from a pool of words in the following categories:

Acoustically similar words (e.g. man, mad, map)

Acoustically dissimilar words (e.g. pen, day ,few)

Semantically similar words (e.g. great, big, large)

Semantically dissimilar words (e.g. hot, old, late)

In the LTM study, each list of words was extended to ten and recall was tested after an interval of

20 minutes.

The findings were that words with similar sounds were much harder to recall using STM than words

with dissimilar sounds. Similarity of meaning had only a very slight detrimental effect on STM.

When participants were recalling from LTM, recall was much worse for semantically similar words

than for semantically dissimilar words. Recall from LTM was the same for acoustically similar and

acoustically dissimilar words.

Baddeley concluded that STM relies heavily on acoustic encoding. LTM primarily makes use of

semantic encoding.

Acoustic encoding in STM

Conrad (1964)

If you have to remember something for a short time, such as a phone number, you probably repeat

it over and over. You repeat it acoustically. What is interesting is that people do this whether they

heard the information (someone told you the number) or saw it (you looked it up in a phonebook).

This suggests that short-term memory may encode information acoustically.

Conrad (1964) investigated this by comparing performance with acoustically and visually presented

data. He presented participants with six letters at a time, displaying each of them for 0.75 of a

second. The participants had to recall the letters in the order that they were presented. When the

letters sounded alike, even though they were visually presented, errors were made in terms of

sound confusions, for example, S was recalled instead of X.

Evaluation of Conrad (1964)

Later research (Posner, 1969) demonstrated that visual codes do in fact exist in STM, at least some

of the time. In Posner’s experiment participants were shown two letters, where the second letter

was either identical to the first letter (e.g. AA), or the same but a different form (e.g. Aa), or different

(e.g. AB). The letters were displayed one at a time, separated by a 2-second interval. Participants had

to say whether the letters were the same or different. Reaction time was longer for Aa than AA

which suggests that participants were processing the data visually rather than acoustically (A and a

sound the same but look different)

Task: As an evaluation point, how do you think that a gadget like the iphone might affect a person’s STM? How might Conrad’s memory experiments be considered to be out-of-date?

Duration of STM

Peterson and Peterson (1959)

The aim of the experiment was to test how long STM lasts when rehearsal is prevented.

The participants were briefly shown a consonant trigram (i.e. 3 letters such as CPW or NGV).

Participants were asked to count backwards in threes from a specified number to stop them

thinking about the letters. After intervals of 3, 6, 9, 12, 15 or 18 seconds, participants were asked to

recall the original trigram. The procedure was repeated several times using different trigrams.

They found that participants were able to recall 80% of trigrams

after a 3-second interval. Progressively fewer trigrams were

recalled as the time intervals lengthened. After 18 seconds, fewer

than 10% of the trigrams were recalled correctly.

Peterson and Peterson concluded that if rehearsal is prevented,

information vanishes rapidly from STM. Therefore, decay is the

mechanism for forgetting in STM.

Evaluation of Peterson and

Peterson (1959)

Trigrams are rather

artificial things to

remember and may not

reflect everyday memory.

It is possible, however,

that interference from the

counting task (not merely

decay) caused the poor

recall.

The experimental method

used in this study allows us

to see the (causal) effect of

time passing (independent

variable) on recall

(dependent variable).

Duration of LTM

Bahrick (1975)

The aim of this study was to establish the existence of a very long-term memory (VLTM), and to see

whether there was any difference between recognition and recall.

For the procedures the investigators tracked down the graduates from a particular high school in

America over a 50-year period. 392 graduates were shown photographs from their high-school

yearbook. For each photo, participants were given a group of names and asked to select the name

that matched the person in the photo (recognition group). Another group of participants were

simply asked to name the people in the photos without being given a list of possible names (recall

group).

Bahrick found that in the name matching condition, participants were 90% correct even 14 years

after graduation. After 25 years, these participants were 80% accurate; after 34 years, 75% accurate

and, even after 47 years, 60% accurate. The 2nd group who had to identify the photos without any

name cues were not quite as successful. They were 60% accurate after 7 years, but the level of

accuracy had dropped to less than 20% after 47 years.

Bahrick concluded that people can remember certain types of information for almost a lifetime. The

accuracy of VLTM is better when measured by recognition tests than by recall tests.

Evaluation of Bahrick (1975) Unlike many memory experiments, this study used meaningful stimulus material (high-school yearbooks) and tested people for memories from their own lives. It is unclear whether the drop-off in accuracy after 47 years reflects the limits of duration or a more general decline in memory with age. TASK: With a learning partner, attempt to assess the methods of this study so that you can find two more evaluation points. Remember to explain and apply each point.

Leading Questions in Eyewitness Testimony (EWT)

Loftus and Palmer (1974)

The study’s general aim was to investigate the accuracy of memory after witnessing a car accident.

In particular it was to see if leading questions distort the accuracy of an eyewitness’s immediate

recall. People are notoriously poor at estimating speed of moving cars and therefore they might be

particularly receptive to any hints (leading questions). This experiment aimed to see if this is true.

The procedures involved 45 students being shown 7 films of different traffic accidents. After each

film the participants were given a questionnaire which asked them to describe the accident and

then answer a series of specific questions about it. There was one critical question: “About how fast

were the cars going when they hit each other?”. One group of participants was given this question.

The other four groups were given the verbs ‘smashed’, ‘collided’, ‘bumped’ or ‘contacted’ in place

of the word ‘hit’.

The mean speed estimate was calculated for each group, as shown in the table. The group given the

word ‘smashed’ estimated a higher speed than the other groups (about 41mph). The group given

the word ‘contacted’ estimated the lowest speed (about 32mph).

Loftus and Palmer concluded that the form of question can have a significant effect on a witness’s

answer. In other words, leading questions can affect the accuracy of memory. Such leading

questions are an example of what psychologists call post-event information – information given

after the event which may alter memory. It is possible that such post-event information causes the

information to be altered before it is stored so that memory is permanently affected. A second

possible explanation is that the form of the question actually alters the participant’s memory

representation of the accident, which leads them to produce a higher or lower estimate.

Many studies have been conducted on EWT, a number of them by Elizabeth Loftus. Here she worked with John Palmer to look at some of the ways that memory can be distorted. Elizabeth Loftus is a prolific researcher in EWT and carrying out some research by yourself may help to improve your skills on your examination. Current information on Loftus can be found on wikipedia: http://en.wikipedia.org/wiki/Elizabeth_Loftus

Evaluation of Loftus and Palmer (1974)

A laboratory experiment may not represent real life because

people don’t take the experiment seriously and/or they are not

emotionally aroused as they would be in a real accident. Foster et

al (1994) found that if participants thought they were watching a

real-life robbery and that their responses would influence the trial,

their identification of the robber was more accurate.

Emotional arousal may actually enhance the accuracy of memory,

as Christianson and Hubinette (1993) found when they

interviewed 110 real witnesses to bank robberies. Those witnesses

who had been threatened were more accurate in their recall and

remembered more details than those who had been onlookers and

less emotionally aroused. This continued to be true even 15

months later.

Evaluation of Loftus and Palmer (1974)

In many experiments, the experimental design leads to certain

inevitable responses from participants. They might feel uncertain

about what to do and how to behave, and would look for cues about

what is expected of them. They would be especially receptive to

certain features of the experiment, such as leading questions. These

features almost demand a particular response and thus these demand

characteristics might explain the findings of Loftus and Palmer’s study.

Loftus and Palmer conducted a second experiment to see if memory was altered by post-event information. A new set of participants was divided into three groups and shown a film of a car accident. Group 1 was given the verb ‘smashed’, group 2 the verb ‘hit’, and group 3 (the control group) was not given any question about the speed of the vehicles. The participants returned one week later and were asked 10 questions about the accident, including another critical question: ‘Did you see any broken glass?’ There was no broken glass in the film but, presumably, those who thought the car was travelling faster might expect that there would be broken glass. The findings are shown in the table on the previous page. 17% of participants in the smashed condition said that they had seen broken glass. This compared to 7% in the hit condition. Participants gave higher speed estimates in the ‘smashed’ condition, as before. They were also more likely to think they saw broken glass. This shows a significant effect of post-event information on later recall of events.

Factors that influence the accuracy of EWT

Wagstaff (2002) argued that factors that can influence the

accuracy of EWT can be remembered using the mnemonic

ADVOKATE. Accuracy is improved under the following

conditions:

Kebbel and Wagstaff (1999) found that witnesses were

normally very accurate when describing a person’s sex,

racial background, style of clothing and hair colour, but

less accurate when describing things like age, height and

overall build.

Loftus (1979) identified the weapon focus effect. There

were two conditions in this experiment. In both conditions

participants heard a discussion in an adjoining room. In

condition 1 a man emerged holding a pen and with

grease on his hands. In condition 2 the discussion was

rather more heated and a man emerged holding a paper

knife covered in blood. When asked to identify the man

from 50 photos, participants in condition 1 were 49%

accurate compared with 33% accuracy in condition 2. This

suggests that the weapon may have distracted attention

from the man and might explain why eyewitnesses

sometimes have poor recall for certain details of a crime.

Yuille and Cutshall (1986) interviewed 13 people who had

witnessed an armed robbery in Canada. The interviews

took place more than 4 months after the crime and

included two misleading questions. Despite these

questions, the witnesses provided accurate recall that

matched their initial detailed reports. This suggests that

post-event information may not affect memory in real life.

This study also shows that EWT can be very reliable.

Amount of time – the

longer the time for which

the event was observed

Distance – the close the

witness is to the event.

Visibility – the clearer the

visibility.

Obstructions – few

obstructions between

witness and event

Known – the more familiar

the person, the better the

identification

Any reason to remember –

the more novel or

emotionally important the

better.

Time – the shorter the time

between the event and the

recall, the better the

memory.

Errors – the more

inaccurate parts of a

witness’s testimony are

sown to be, the less reliable

is the rest of their

testimony.

Factors affecting accuracy of Eyewitness testimony (EWT)

List (1986)

List applied the idea of schemas to EWT by drawing up a list of elements that might occur in a typical

shoplifting scenario. She asked people to rate these events in terms of how likely they were to occur

in a typical shoplifting incident. She then compiled a video showing eight different shoplifting

incidents and included some elements that people had rated as high probability and some that

people had rated as low probability. She then showed the video to a new set of participants and

asked them, a week later, to recall what they had seen. She found that participants were more likely

to recall high probability events than low probability events and that they often reported seeing high

probability elements that had not actually been included in the video at all.

Methodological issues

This was a laboratory-based experiment, but List took trouble to try to make her video realistic. This

is, however, not the same as a real-life incident where witnesses would be expecting anything to

happen.

List used a pilot study to find out what elements people most commonly associate with shoplifting.

Ethical issues

As with any study, List needed to obtain the consent of her participants and to debrief them

afterwards.

Factors affecting accuracy of EWT – age of witness

Poole and Lindsay (2001)

Poole and Lindsay engaged children aged 3 to 8 in a science demonstration. The parents of the

children then read them a story, which contained some of the elements of the science

demonstration but also included novel information. The children were then questioned about the

science demonstration and it was found that they had incorporated much of the new information

(i.e. from the parents’ story) into their original memory. In another phase of the experiment, the

children were asked to think very carefully about where they had got their information from (this is

called source monitoring) and some of the older children then revised their account of the science

demonstration and extracted the post-event information. However, the younger children did not

seem able to do this. This has important implications for measuring the accuracy of small children’s

testimony since they seem very poor at source monitoring.

Methodological issues

This was an experiment but more difficult to eliminate extraneous variables than one using artificial

stimuli in a highly controlled laboratory setting.

Investigators have to be particularly careful when using children, particularly those as young as 3 and

4 to make sure they understand instructions and that they are paying attention.

Ethical issues

There are particular factors to be taken into account when using young children who may not be

able to give informed consent on their own behalf. Parents must give informed consent but, in this

case, it was helpful that the parents were involved as well and so the children were with familiar

people and less susceptible to investigator effects.

Misleading information and the use of the cognitive interview

Fisher et al (1989)

The aim of this study was to further test the validity of the cognitive interview technique using a

field experiment. Precious research had mainly used laboratory experiments and the cognitive

interview seemed highly effective in such studies.

16 experienced police inspectors from Miami, Florida, conducted two interviews on 47 witnesses

or victims of shoplifting or mugging. Between the two interviews, 7 of the police officers were

trained in using the cognitive interview technique. The other 9 officers formed the control group

(they received no further training). The researchers measured both the increase in facts elicited in

the 2nd interview compared to the 1st, and the number of facts elicited by the police officers in the

cognitive interview group compared to the control group. The independent variable was the type of

interview the police officers used on the 2nd interview (cognitive or standard) and the dependent

variable was the number of accurately recalled facts produced by each of these techniques.

Results provided overwhelming support for the cognitive interviewing technique. The cognitive

interviewers obtained 47% more facts relating to matters already examined, whereas there was no

gain from the 2nd interview for the control group.

Cognitive interviews are a useful technique for improving EWT when compared to a standard

interviewing technique. This seems to occur

both in a laboratory and more

realistic field experiments.

The officers in the control

group were aware that they

were not being given any

extra interview training and

this might have affected

their motivation levels. It

may have been this factor

that led to the difference

in recall accuracy

between the

experimental and

control groups

Evaluation of Fisher et al (1989)

Practical problems involving application of cognitive interviewing: some research has

suggested problems with cognitive interviews. Memon et al (1994) failed to find an

improvement using a cognitive interview technique when used by police officers. In their

study, 38 experienced police officers were trained for 4 hours on general interviewing

techniques. Half of the police officers were then trained in cognitive interviewing

techniques. Immediately after this, the police officers were asked to interview witnesses of a

staged event (a simulated armed robbery in the parking lot of the police training school a few

hours earlier). The results showed no benefits of cognitive interviewing over standard

interviewing in terms of the number of correct answers or the number of incorrect answers

provided by the witnesses. Moreover, it appeared as though the cognitive interviewing

training had little effect on the interview procedures used by the police officers, so perhaps

the lack of any effect may have been due to either the training or a reluctance on the part of

the officers to fully implement the training.

Limited use by the police: there are concerns among the police that the use of the ‘change of

perspective’ mnemonic may mislead witnesses into thinking that they are being asked to

speculate on the event they witnessed. After all, no one can be certain what another person

saw. Due to worries such as these, this particular mnemonic is less frequently used in

practice (Memon et al, 1993).

Types of attachment

Ainsworth and Bell (1970)

The aim of this study was to produce a method of assessing quality of attachment by placing an

infant in a situation of mild stress (to encourage the infant to seek comfort) and of novelty (to

encourage exploration behaviour). Both comfort-seeking and exploration behaviour are indicators of

quality of attachment.

The procedures involved about 100 middle-class American infants and their mothers taking part in

the study. A method of controlled observation was developed. This involved observing infants with

their mothers during a set of predetermined activities (this is known as the Strange Situation). All of

the sessions, apart from the first one, lasted three minutes.

Observers recorded the infants’ and mothers’ behaviours, especially noting the following:

Separation anxiety: the unease the infant showed when left by the caregiver

The infant’s willingness to explore

Stranger anxiety: the infant’s response to the presence of a stranger

Reunion behaviour: the way the caregiver was greeted on return

The observational record led Ainsworth and Bell to classify the infants into 3 broad groups:

Type B: Securely Attached (66%) – One group of infants tended to explore the unfamiliar

room; they were subdued when their mother left and greeted her positively when she

returned. The infants showed moderate avoidance of the stranger, although were friendly

when mother was present. The mothers were described as sensitive.

Type A: Avoidant-insecure (22%) – A second group did not orientate to their mother while

investigating the toys and room; they did not seem concerned by her absence and showed

little interest in her when she returned. These infants also avoided the stranger, but not as

strongly as they avoided their mother on her return. It was observed that these mothers

sometimes ignored their infants.

Type C: Resistant-insecure (12%) – A third group showed intense distress, particularly when

their mother was absent, but they rejected her when she returned. These infants showed

ambivalent behaviour towards the stranger, similar to the pattern of resistance and interest

shown to their mother on her return. These mothers appeared to behave ambivalently

towards their infants.

The researchers concluded that there are significant individual differences between infants. It also

shows that most American children are securely attached. There appears to be a distinct association

between the mothers’ behaviour and the infants’ attachment type, which suggests that mothers’

behaviour may be important in determining attachment type.

Evaluation of Ainsworth and Bell (1970)

It would be unreasonable to make generalisations about all

infant behaviour on the basis of this sample. The study and

its findings are restricted to middle-class American infants

i.e. are culturally biased

In another study, Main and Cassidy (1988) identified a

further group of children; this classification group is referred

to as disorganised (type D). These children show

inconsistent behaviour, confusion and indecision. They also

tend to freeze or show stereotyped behaviour such as

rocking.

A father with his child – someone who is ignored (i.e. not studied) on the Strange Situation

Development of Attachment

Schaffer and Emerson (1964)

The researchers designed a large-scale longitudinal study to find out more information about the

development of attachment. Over a period of two years they followed 60 infants from a mainly

working class area of Glasgow, keeping a detailed record of their observations. The infants were

observed every four weeks until they were 1 year old and then again at 18 months. At the start of

the investigation the youngest participant was 5 weeks and the oldest was 23 weeks.

Attachment was measured in two ways:

Using separation protest in 7 everyday situations. The infant was left alone in a room, left

with other people, left in his/her pram outside the house, left in his/her pram outside the

shops, left in his/her cot at night, put down after being held by an adult, or passed by while

sitting in his/her cot or chair.

Using stranger anxiety. Every visit started with the researcher approaching the infant and

noting at what point the infant started to whimper, thus displaying anxiety.

Separation protest and stranger anxiety are signs that an attachment has formed. Before this stage

of specific attachments infants show neither of these behaviours.

Schaffer and Emerson found that:

Half of the children showed their first specific attachment between 25 and 32 weeks (6-8

months). Fear of strangers occurred about a month later in all the children.

The intensity of attachment peaked in the first month after attachment behaviour first

appeared, as measured by the strength of separation protest. However there were large

individual differences. Intensely attached infants had mothers who responded quickly to

their demands (high responsiveness) and who offered the child the most interaction. Infants

who were weakly attached had mothers who failed to interact.

Multiple attachments: soon after one main attachment was formed, the infants also

became attached to other people. By 18 months very few (13%) were attached to only one

person, 31% had five or more attachments, such as father, grandparent, or older sibling. In

65% of the children the first specific attachment was to the mother, and in a further 30% the

mother was the first joint object of attachment.

Time spent with infant: in 39% of the cases the person who usually fed, bathed, and

changed the child was not the child’s primary attachment object. In other words, many of

the mothers were not the person who performed these tasks yet they were the main

attachment object.

Evaluation of Schaffer and Emerson (1964) This study remains one of the largest longitudinal studies of infant attachment behaviour and a classic, despite the fact that it was conducted more than 40 years ago. Criticisms can be made about the methodology, perhaps most importantly about some of the data collection. The data were collected either by direct observation or from the record kept by the mothers. These are both prone to bias and inaccuracy. Mothers were asked to record situations where separation protest was shown, and to whom these protests were directed. It is quite possible that a busy mother may have had to manufacture these records some days after the events and her memory would have been influenced by expectations. On the other hand, such data would have been more accurate than the retrospective recollections used in many studies, and would have more ecological validity that data collected in laboratory observations. (e.g. the Strange Situation)

Learning theory of Attachment

Harlow (1959)

Harry Harlow was initially interested in the psychology of learning, and conducted experiments using

rhesus monkeys. In order to reduce mortality rates in the experimental participants, Harlow found it

best to separate them from their mothers after birth and raise them in sterile conditions. The

monkeys were provided with a nappy on the floor of their cages because it had been observed that

the monkeys liked to touch these, especially when feeding from a bottle. In fact, the monkeys

appeared to become extremely attached to these nappy pads as indicated by the fact that they

became very agitated when the pads were removed daily for cleaning.

This attachment reminded Harlow of a child’s fondness for a special blanket or a cuddly toy, and

suggested to him an idea for an experiment. Would the monkeys prefer the activity of feeding to

that of bodily comfort? To test this Harlow (1959) arranged for very young rhesus monkeys to be

taken from their mothers and placed in cages with two surrogate (or substitute) mothers, as shown

in the picture. One of the “mothers” was made of wire and the other was covered in cloth. Milk was

provided by the wire mother for some of the monkeys, whereas it was provided by the cloth

mothers for the others. The findings were clear cut. The monkeys spent most of their time on their

cloth mother even when she did not supply milk.

The cloth mother provided “contact comfort” which was clearly preferable. If the monkeys were

frightened they ran to their cloth mother. However the monkeys did not develop into normal adults.

Later in life the monkeys were either indifferent or abusive to other monkeys and had difficulty

Evaluation of Harlow (1959)

Monkeys are social animals who usually live in quite large

groups. This means that Harlow’s studies could be criticised on

two grounds: modern ethical considerations would make this

experimental approach hard to justify, but the studies could

also be methodologically flawed.

The rhesus monkeys were reared in isolation. These are

monkeys who come from a highly social background. After the

experiments were completed, the monkeys exhibited severely

disturbed behaviour. At the time, the distress caused to the

monkeys was seen as acceptable in the light of possible

benefits of understanding how attachment develops.

However, it is unlikely that such a study would be allowed to

proceed today because of stricter ethical guidelines for

research using non-human participants.

The baby monkeys were doubly deprived of both maternal

care and the company of others. Therefore we cannot be

certain whether their maladjustment was due to maternal

deprivation or social deprivation, or both.

with mating and parenting. This shows that contact comfort is preferable but not sufficient for

healthy development. Presumably infants need a responsive caregiver and an interactive

relationship with this individual for healthy development.

Harlow conducted various further studies to investigate the effects of deprivation. Harlow and

Harlow (1962) raised monkeys for lengthy periods in total isolation. When they were placed with

other monkeys they remained withdrawn and extremely fearful. In comparison, monkeys raised

with a cloth “mother” were much more able to engage in social activity. This shows that the cloth

mother was better than nothing.

In another experiment four young monkeys were raised on their own, without any “mother”. They

spent the first few months huddled together but gradually developed more independence and

finally appeared to have suffered no ill effects. This suggests that the infant-infant affectional bond

can be just as effective as mother-infant.

Disruption of Attachment – the effects of deprivation

Bowlby (1944)

In the 1930’s John Bowlby worked as a psychiatrist in a Child Guidance Clinic in London, where he

treated many emotionally disturbed children. Part of this treatment would involve interviews with

parents about the child’s early experiences. This work led Bowlby to formulate his belief about the

link between early separations and later maladjustment.

The aim of this study was to test the maternal deprivation

hypothesis. Bowlby was specifically interested in one group of

adolescents attending the clinic – the ‘thieves’. All the children

referred to the clinic were emotionally maladjusted but, Bowlby

argued, some of them were unable to understand how other people

felt. These children lacked normal signs of affection, shame or a sense

of responsibility. Such characteristics enable them to be ‘thieves’ –

they could steal from others because it didn’t matter to them. So

Bowlby decided to compare 44 juvenile thieves (children referred to

the clinic because they had been involved in thefts) with another

group of emotionally disturbed teenagers who were not thieves.

Were ‘thieves’ more likely to have experienced early separations during which they received little

substitute care? This would suggest that early separations could lead to emotional maladjustment,

supporting the maternal deprivation hypothesis.

The children interviewed ranged in age from 5 to 16. They all attended the Child Guidance Clinic. The

experimental group consisted of 44 children who had been referred to the clinic because they were

caught stealing. A further 44 children acted as a control group. This group were not thieves but had

experienced emotional problems, for which they had been referred to the clinic. In addition Bowlby

diagnosed 14 of the thieves as affectionless psychopaths – individuals who particularly lacked

emotional sensitivity. The children and their parents were interviewed by Bowlby and his associates.

Information was collected about their early life experiences, with particular attention paid to early

separations.

It was found that the affectionless thieves had experienced frequent early separations from their

mothers. The figures in the table show that 86% of affectionless thieves (12 out of 14) experienced

frequent separations compared to 17% (5 out of 30) of the other thieves. Furthermore, almost none

of the control participants experienced early separations, whereas 39% of all the thieves had

experienced early separations. These early separations often consisted of continual or repeated

stays in foster homes or hospitals, when the children were often visited by their families.

These data suggest a link between early separation and becoming a thief (i.e. someone lacking a

social conscience). They also suggest that early separations are linked to affectionless psychopathy.

In other words, lack of continuous care may well cause emotional maladjustment, especially in the

extreme form of affectionless psychopathy.

Evaluation of Bowlby (1944)

Unreliable data – the information about early separation was collected

retrospectively. Parents were asked to recall events from up to 14 years previously

about when they had been separated from their children. It is likely that such recall

was not completely accurate. The data collected may also have been unreliable

because people don’t always answer interviewer’s answers truthfully – parents

might be expected to present themselves in a better light (social desirability bias)

and report fewer separations.

Correlation, not a cause – this study was not an experiment – it involved

correlational analysis. It is wrong to think that early separations caused later

maladjustment in the children just because the two are linked together. Rutter

(1981) suggested that there may be a factor which causes both separations and

maladjustment and that is why they are linked. A possible factor would be stress in

the home due to marital discord or a lack of resources to care for the children.

Psychologists have to take great care to avoid confusing a correlation with a cause.

Failure to form attachment and the effects of institutionalisation

Hodges and Tizard (1989)

The aim of the study was to investigate the effects of early privation on subsequent social and

emotional development, and to test the maternal deprivation (or privation) hypothesis. The study

aimed to follow the same children over a long period of time (a longitudinal study) to collect reliable

information linking early experiences to later outcomes for the same individuals.

This longitudinal study was a natural experiment. The independent variable (attachment experience)

varied naturally. The participants were 65 children who had been placed in an institution when they

were les than 4 months old. There was an explicit policy in the institution against caregivers forming

attachment with the children. This would suggest that the children experiences early privation. By

the age of 4, 24 of the institutionalised children had been adopted, 15 had returned to their natural

homes, and the rest remained in the institution. Assessment at ages 8 and 16 involved interviewing

those children who were adopted and those who had returned to their original homes. Their parents,

their teachers and their peers were also interviewed. Data were also obtained about a control group

of ‘normal’ peers.

There were some differences between the adopted and ‘restored’

children. The adopted children generally had close attachments to

their parents and good family relationships, whereas this was

much less true for the restored children. However, there were

similarities in the behaviour of the adopted and restored children

outside the family. For example, both groups were more likely to

seek adult attention and approval than control children, and both

groups were less successful in peer relationships.

There is evidence that does not support the maternal deprivation hypothesis. The two ex-institution groups, adopted and restored, differed within their family relationships. The restored children often returned to the same difficult circumstances that had precipitated the need for care in the first place, and to parents who may have felt ambivalent about them. In contrast, adopted children went to homes where the parents had very much wanted a child. This shows that recovery is possible under the right circumstances. There is evidence that does support the maternal deprivation hypothesis. Outside the family environment it would appear that early privation did have an effect on subsequent social development. Clarke and Clarke (1979) put forward a transactional model to explain the findings. It may be that the adopted children in Hodges and Tizard’s studies got on well within their families because the families made special efforts to love them, whereas they did not experience this outside the home and thus were unable to form relationships as easily or well.

Evaluation of Hodges and Tizard (1989)

Random allocation of participants to experimental groups is used to ensure that the

participant groups in an experiment are equivalent. In Hodges and Tizard’s study, there may

have been important differences between the two groups – the adopted and the restored

group – besides the independent variable. It is possible, for example, that the children

selected for adoption were the more attractive and socially able children. The children’s

temperament thus becomes a confounding variable in this study – confounding because it

acts as another independent variable. This means that we cannot infer a causal relationship

between the effects of early privation on later social and emotional development from this

study.

Attrition is a common problem in longitudinal research. Inevitably, some participants are no

longer available or willing to take part in the study as the years pass by. It is possible that a

certain kind of individual is more likely to drop out from the study – for example, those who

are less highly motivated or, in the case of Hodges and Tizard’s study, those who are less well

adjusted. This leaves the study with a biased sample. If a study sample is biased, the

researchers have to be very careful about what conclusions they can draw and it is not

appropriate to generalise the findings.

Failure to form attachment and the effects of institutionalisation.

Rutter et al (2007)

Rutter et al (2007) are carrying out an ongoing longitudinal study, comparing Romanian orphans

who were adopted by UK families with UK-born adoptees who were placed with families before they

were six months of age. The Romanian adoptees entered the orphanage as small babies between

one and two weeks old. Conditions in the institutions were very poor. 58 babies were adopted

before they were 6 months old and 59 were adopted between 6 and 24 months of age. 48 babies

were classed as late placed adoptees (between two and four years of age). These formed the three

conditions of the naturally-occurring independent variable. At the time of adoption, over half of the

Romanian children showed evidence of severe malnourishment. They were in the bottom third of

the population for weight and head size. Some of these children have been followed up at ages four,

six and eleven years, using a range of measures including interviews and observations of the child’s

behaviour.

At age six, Rutter et al found evidence of disinhibited attachment which they defined as ‘a pervasive

pattern of attention seeking behaviours with a relative lack of selectivity in social relationships’.

No disinhibition Mild disinhibition Marked disinhibition

UK adoptees 21 (40.4%) 29 (55.8%) 2 (3.8%)

Romanian adopted

(before 6 months)

24 (53.3%) 17 (37.8%) 4 (8.9%)

Romanian adopted

(6-24 months)

26 (29.5%) 39 (44.3%) 23 (26.1%)

As the table shows, disinhibited behaviour was most common in the late adopted Romanian group

with over one-quarter (26.1%) showing ‘marked’ disinhibited attachment behaviours. The behaviour

pattern was extremely rare in UK adoptees (3.8%) and early adopted children (8.9%). This suggests

that disinhibited attachments are more likely in children who have experienced longer periods in

institutions.

In 2007, some of these children were followed up aged 11. Rutter found that the disinhibited

behaviour pattern had persisted in many adoptees. Of the 83 Romanian children showing mild or

marked disinhibited attachment at age 6, 45 (54%) of these still showed this five years later. While

this indicates a drop in the number of children showing this behaviour pattern at age 11, it is still

well over half. Even more worryingly, Rutter found that many of the children showing disinhibited

attachments were receiving help from either special educational and/or mental health services.

● ● ●

Methodological issues

This study is using a range of measurements to assess the children’s behaviour including semi-

structured interviews and observation to see if the child makes inappropriate physical contact.

This makes the research rich and detailed.

As with Tizard and Hodges’s study, participant attrition is an issue in this longitudinal research

study.

Rutter et al acknowledge that it has been difficult to obtain information about the quality of

care in many of the institutions in Romania making it difficult to assess the extent of privation

in the early environments of the children in the study.

Ethical issues

As with other longitudinal studies in this area, researchers need to be sensitive to the needs of

both the children and their adoptive families in research of this nature. The extent of

participant attrition shows that some families may wish to remove themselves from further

study and to bring up their children outside the glare of research. Freedom to withdraw from

a research study is an important ethical principle.

● ● ●

Cultural variations in attachment

Takahashi (1990)

The aim of this study was

to consider whether it is

appropriate to use the

Strange Situation with

Japanese children. The key

question is whether the

Strange Situation is a valid

procedure for cultures

other than the original one

i.e. other than American,

middle-class, white, home-

reared infants and their

mothers. By making

comparisons between the

American and Japanese

group, it may be possible to reveal the cultural assumptions on which the procedure is based. The

participants were 60 middle-class Japanese infants, aged 1 year, boys and girls, and their mothers.

They were all raised at home. The infants and mothers were observed in the Strange Situation.

68% of the infants were classified as securely attached, almost identical to the original American

sample. There were no infants classified as avoidant-insecure. 32% were classified as resistant-

insecure. When the observational data were examined in more detail, differences emerged. The

Japanese infants were much more disturbed after being left alone. In fact, the ‘infant alone’ step

was stopped for 90% of the participants because the infants were so distressed. If the infants had

not become so distressed, many more of them (possibly more than 80%) would have been classified

as securely attached (the observation of distress led to an alternative classification).

The findings suggest that there are cross-cultural variations in the way infants respond to

separation and being left alone. This difference may be due to the fact that Japanese infants are

almost never left alone. This means that, for them, the Strange Situation was more than mildly

stressful, and thus the behaviours observed were reactions to extreme stress – not the original aim

of the Strange Situation. The findings also highlight a second cross-cultural variation – the total lack

of avoidant behaviour in this sample. This can also be explained in cultural terms. Japanese children

are taught that such behaviour is impolite and they would be actively discouraged from displaying it.

The final conclusion must be that the Strange Situation does not have the same meaning for the

Japanese as it does for American participants and therefore is not a valid form of assessment for

that culture.

Evaluation of Takahashi (1990)

Research with children, especially with infants, needs to be especially careful in terms of

potential psychological harm to participants, which is an important ethical issue.

Takahashi’s study was carried out on a limited sample of only middle-class, home-reared

infants (as was the original study by Ainsworth and Bell). It may therefore not be

reasonable to generalise these findings to all Japanese people, although the results do

demonstrate that there are important cultural or sub cultural differences in attachment.

Task:

With a learning partner, assess the key evaluation

points of Takahashi’s study.

Can you find more strengths and weaknesses?

Remember to state, explain and apply each of your

points.

Cultural variations in attachment

Van Ijzendoorn and Kroonenberg (1988)

Van Ijzendoorn and Kroonenberg wanted to compare the findings of studies using the Strange

Situation conducted in different cultures. They were interested in patterns of attachment type

(secure/avoidant/resistant) both between and within cultures.

They reviewed 32 worldwide studies involving 8 countries and over 2,000 children. These studies

had been conducted by other researchers, and Van Ijzendoorn and Kroonenberg compared the

studies, looking for any general trends.

Country No. of studies Securely attached Anxious-Avoidant Anxious-Resistant

West Germany 3 57% 35% 8%

Great Britain 1 75% 22% 3%

Netherlands 4 67% 26% 7%

Sweden 1 74% 22% 4%

Israel 2 64% 7% 29%

Japan 2 68% 5% 27%

China 1 50% 25% 25%

USA 18 65% 21% 14%

The overall worldwide pattern was similar to Ainsworth et al’s ‘standard’ pattern. This was also true

of the 18 American studies as a whole, although there was considerable variation within these

studies. The standard pattern shows Type B to be the most common and Type C the least common.

However, Type A is the least common in Israel and Japan, and Type C is relatively more common. But

there were marked differences between the two Japanese studies. In one there were no Type A

children at all, but a high proportion of Type C; in the other, the pattern was more like that of

Ainsworth et al. So, there are marked differences within cultures – as well as between some.

Evaluation of Van Ijzendoorn and Kroonenberg

The countries included in the review are both Western and non-Western (China and Japan),

but only 3 of the 32 studies were carried out in China and Japan. The different patterns of

reaction to the Strange Situation seem to reflect cultures’ values and practices. For example,

the greater frequency of Type A (anxious-avoidant) in Germany may reflect cultural emphasis

on early independence training. Rogoff (2003) cites a study by True et al (2001) involving the

Dogon people of West Africa, which found a complete absence of Type A; this may stem from

the community’s infant care practices, which involve responsiveness, constant closeness to

mothers, and immediate nursing in response to signs of stress.

The greater frequency of Type C in Japan may result from greater stress during the Strange

Situation due to infants’ unfamiliarity with being left with strangers. Japanese children are

rarely separated from their mothers, so the separation episodes are the most upsetting for

these children. In contrast, for African-American infants, who are used to being tended by

several caregivers and who are encouraged to be friendly to the many strangers they

encounter; the Strange Situation may arouse their interest in exploration (Rogoff, 2003). For

children raised on Israeli kibbutzim (small, close-knit communities), it’s the arrival of the

stranger that’s most distressing. This shows that the Strange Situation has a different meaning

in different cultures; we need to know about the child-rearing practices of different cultures if

we’re to interpret findings based on the Strange Situation validly.

Implications of research into attachment and day care for child care practices

Andersson (1992)

Early research into the effects of day care focused on the question of whether it was a good or bad

thing. As research progressed psychologist realised that this question was too simplistic. Research

should focus on what features of the day care situation lead to positive experiences and what factors

lead to negative outcomes. More recent research has also looked at the long-term effects of the day

care experience as opposed to more immediate outcomes.

One example of this is a longitudinal study by Andersson (1992) conducted in Sweden. A sample of

over 100 children were studied from both lower and middle-class homes as well as one-parent

families. The families were first contacted when the children were aged between 3 and 4. At this

time their early child-care arrangements were recorded, most importantly the age that they first

started in day care. The children were assessed at age 8 and 13. Cognitive and socio-emotional

competence was rated by their classroom teachers and IQ test data were collected. Andersson

found that school performance was rated highest in those children who entered day care before

the age of 1. School performance was lowest for those who did not have any day care. This suggests

that day care is not harmful in terms of development and may even be beneficial. There is one

important factor. Those children who did enter day care before the age of 1 also came from families

with higher socio-economic status. This means that the reason such children did better was possibly

related to their socio-economic status (better-off families are better educated and produce more

well-adjusted children). Nevertheless the day care experience could have been detrimental, but it

was not.

Evaluation of Andersson (1992)

It may be significant that this study was conducted in Sweden. A similar

study looking at children in Texas (Vandell & Corasaniti, 1990) found that

children with extensive child-care experiences from infancy were rated by

parents and teachers as having poorer peer relationships and emotional

health. The different results may be due to the different infant-care

practices in both cultures. In Sweden a substantial part of in the infant’s first

year is spent with one or both parents due to a special parental leave

system. This means that infants may be able to form stronger attachments

at home. In addition day care in Sweden is given a great deal of financial

support from the government and therefore may be of a generally higher

quality that in the US.

Andersson stresses this factor, that in Sweden day care has very low

staff/child ratios and carers are highly trained. Texas, on the other hand, is

a state with very low official requirements for child-care facilities. This

emphasises the point that it is quality of care that should concern us, rather

than the question of whether or not day care is a good thing.

Research on the

positive effects

of day care on

children’s social

development

Vandell et al (1988)

They reported that children in

better-quality day care have more

friendly and fewer unfriendly

interactions than those in lower-

quality day care.

Schweinhart et al (1993)

Investigated day-care programs for

disadvantaged children. They

conducted follow-up studies of 66

children involved in the High/Scope

Perry Preschool Project. They found

decreased rates of self-reported

delinquency at age 14, decreased

official chronic delinquency at age 19,

and in the most follow-up at age 27,

lower adult criminality

Creps and Vernon-Feagans (1999)

They found that infants who started

day care before the age of 6 months

were more sociable later than those

who started later. This suggests that

early day care may be better for social

development.

Task: Your AS level examination requires you to discuss the impact of different forms of

day care on children’s social development (including effects on aggression and peer

relations). For each of above research studies, find a strength of the method/finding and

a weakness of the method/finding. Remember to explain and apply your point.

Research on the

negative effects

of day care on

children’s social

development

Belsky & Rovine (1988)

They assessed attachment (using

the Strange Situation) in infants

who had been receiving 20 hours

or more of day care per week

before they were 1 year old. These

children were found to be more

likely to be insecurely attached

compared with children at home.

A recent NICHD report (2001)

concluded that children who

are separated from their

mothers for more than 10

hours a week early in life are

more aggressive once they

reach kindergarten, as rated by

their mothers and by their

teachers.

Violata & Russell (1994)

They did a meta-analysis of the

findings of 88 studies, concluding

that regular day care for more

than 20 hours per week had an

unmistakably negative effect on

socio-emotional development,

behaviour and attachment of young children.

Task: Your AS level examination requires you to discuss the impact of different forms of

day care on children’s social development (including effects on aggression and peer

relations). For each of above research studies, find a strength of the method/finding and

a weakness of the method/finding. Remember to explain and apply your point.

Experimental Method Definition of a Field experiment Example of a Field experiment Strengths of a Field experiment

Weaknesses of a Field experiment

Research Methods – Methods and

Techniques

Experimental Method Definition of a Laboratory experiment Example of a Laboratory experiment Strengths of a Laboratory experiment

Weaknesses of a Laboratory experiment

Experimental Method Definition of a Natural/Quasi experiment Example of a Natural/Quasi experiment Strengths of a Natural/Quasi experiment

Weaknesses of a Natural/Quasi experiment

Use this space to draw

your scatter graphs:

Correlational Analysis Definition of Correlation Example of Correlation Strengths of Correlation

Weaknesses of Correlation

What is a Correlation Coefficient? For the following coefficients, explain the type (e.g. positive or negative) and strength (e.g. strong or weak) of the correlation.

+0.88 -0.35 -0.99 +0.55 +0.21 -0.65

Illustrate all 6 correlation coefficients using scatter graphs

Questionnaires Opinion Surveys Definition of an opinion survey Example of an opinion survey Strengths of an opinion survey

Weaknesses of an opinion survey

Psychological Tests

Definition of a Psychological test

Example of a Psychological test

Strengths of a Psychological test

Weaknesses of a Psychological test

Case Studies Definition of a Case Study Example of a Case Study Strengths of a Case Study

Weaknesses of a Case Study

Interviews Definition of a Structured Interview Example of a Structured Interview Strengths of a Structured Interview

Weaknesses of a Structured Interview

Definition of a Semi-Structured Interview Example of a Semi-Structured Interview Strengths of a Semi-Structured Interview

Weaknesses of a Semi-Structured Interview

Definition of a Clinical Interview Example of a Clinical Interview Strengths of a Clinical Interview

Weaknesses of a Clinical Interview

Definition of an Unstructured Interview Example of an Unstructured Interview Strengths of an Unstructured Interview

Weaknesses of an Unstructured Interview

Observational method Definition of Controlled Observation Example of a Controlled Observation Strengths of a Controlled Observation

Weaknesses of a Controlled Observation

Definition of a Naturalistic Observation (Obs) Example of a Naturalistic Obs Strengths of a Naturalistic Obs

Weaknesses of a Naturalistic Obs

Definition of a Participant Observation Strengths of a Participant Observation

Example of a Participant Observation

Weaknesses of a Participant Observation

Research Methods – Investigation

Design

What is a Variable? What is an Operationalised Variable? What is an Independent Variable? What is a Dependent Variable?

What is a Hypothesis? What is a Directional (1-tailed) hypothesis? What is a Non-Directional (2-tailed) hypothesis? What is a Correlational hypothesis? What is a Null hypothesis?

What is an Aim? Give an example of an Aim

What is an extraneous (confounding) variable?

Task: Write an example of: 1. A directional hypothesis 2. A non-directional hypothesis 3. A correlational hypothesis 4. A null hypothesis Label the IV and DV in each hypothesis.

Examples of Hypotheses:

What is a Repeated Measures design? What is an Independent Measures design? What is a Matched Pairs design?

Strengths of Repeated Measures design

Weaknesses of Repeated Measures design

Strengths of Independent Measures design

Weaknesses of Independent Measures design

Strengths of Matched Pairs design

Weaknesses of Matched Pairs design

What is Sampling? What is Random Sampling? What is Opportunity Sampling? What is Self-Selecting Sampling? What is Stratified Sampling?

Strength of Random Sampling

Weakness of Random Sampling

Strength of Opportunity Sampling

Weakness of Opportunity Sampling

Strength of Self-Selecting Sampling

Weakness of Self-Selecting Sampling

Strength of Stratified Sampling

Weakness of Stratified Sampling

Definition of Face validity Definition of Concurrent validity Definition of Predictive validity

What is a situational variable? How can you control a situational variable? What is a participant variable? How can you control a participant variable?

What is reliability? What is validity? Definition of internal validity Definition of external validity

British Psychological Society (BPS) guidelines for research with human participants http://www.bps.org.uk/the-society/code-of-conduct/ethical-principles-for-conducting-research-with-human-participants.cfm#principles Consent Deception Debriefing Withdrawal from investigation Confidentiality Protection of participants Observational research Giving advice Colleagues Colleagues

Dealing with deception Debriefing: Retrospective informed consent:

Dealing with protection of participants Right to withdraw: Terminating research: Debriefing:

Dealing with informed consent Prior general consent: Presumptive consent: Children as participants:

What is qualitative data? Give an example of qualitative data: What is quantitative data? Give an example of quantitative data

Measure of central tendency What is the mean? What is the median? What is the mode?

Strength of the mean:

Weakness of the mean:

Strength of the median:

Weakness of the median:

Strength of the mode:

Weakness of the mode:

Research Methods – Data analysis and

presentation

Measures of dispersion What is the range? When should it be used? Strength of the range:

Weakness of the range:

Measures of dispersion What is the standard deviation? When should it be used? Strength of the standard deviation:

Weakness of the standard deviation:

What is content analysis? Give an example of content analysis

What is pure qualitative analysis? Give an example of pure qualitative analysis

What are the features of a histogram? Draw an example of a histogram:

What are the features of a bar chart? Draw an example of a bar chart:

What are the features of scattergram? Draw an example of a scattergram: