Hancock - individual differences

• What are the assumptions of the individual differences area?

  - to understand behaviour we should focus on differences rather than what we have in common

  
  

  -We should measure these differences using psychometric testing
• what is the key theme of Hancock's study

  Measuring differences
• What is PCL-R


  -Usually completed by a trained professional during interview with the individual

  
  

  - Hare determined 30 to be a

  threshold after which the individual is deemed a psychopath.
• What are clinical characteristics of a psychopath

  manipulative,

  grandiose

  sense of self-worth, pathological lying,

  lack of

  remorse, lack of empathy, shallow emotions, parasitic lifestyle.
• 3 language characteristics


  unique socioemotional needs (food, money, sex, etc.), poverty of affect (past tense

  verbs, e.g. stabbed; disfluencies e.g. erm, uh...), instrumental and world predatory view (that, because, in order to get X,

  I had to do Y').
• Oxman 1988

  -Statistical speech analysis shown to be more accurate than a clinician's diagnosis.
• Porter (2009)

  -popular notion that psychopaths are skilled conversationalists - indeed, Porter found that

  psychopathic offenders in the Canadian penal system were approx. 2.5x more likely than non-psychopaths to be

  successful in their parole applications, despite being far more likely to reoffend.
• Williamson 1993

  psychopathic language might actually be less cohesive and more incoherent than that of non-

  psychopaths.
• What was the aim of Hancocks study ?

  to compare the crime narratives of psychopaths compared with non-psychopathic murderers

  
  

  -speech analysed for instrumental worldview, unique socioemotional needs and poverty of affect.
• What was the method of H's study?

  - self-report method- semi-structured face-to face interviews

  
  

  (used Step-Wise Interview Technique)

  
  

  > Potentially a quasi-experiment (IV - psychopath or not; DV - number of language characteristics

  used).

  >Psychopathy was measured using Psychopathy Checklist Revised (Hare)
• What was the strength of the method (self-report)

  [P] More detailed info obtained from each participant as subsequent questions

  specifically shaped to the participant

  
  

  [E] for example participants could be asked to elaborate on the crime

  they committed, etc.,

  
  

  [C] therefore increasing validity
• What was a weakness of the self-report method?

  [P] Considering how time-intensive they are, it means that there will be less participants available,

  [E] in

  this case only 14 psychopaths were interviewed

  [C] potentially reducing population validity.
• What was the sample of Hancocks study ?

  - self-selected sample of 14 psychopaths + 38 non-psychopaths

  (all male - androcentric)

  
  

  - all from Canadian correctional facility

  
  

  -mean age at time of homicide 28.9 years

  
  

  similar mean time since crime 11.87 P vs 9.82 Non-p
• What was the procedure of Hancocks study ?

  PCL-R

  - 20 criteria scored from 0 - 2 for a maximum score of 40

  clinical diagnostic cut-off for psychopathy is

  scores of 30 or above, however the researcher used 25.

  - Conducted by trained prison psychologists or a researcher who trained in the coding of the PCL-R.

  Audiotaped interview (25 mins)

  - The aim (to examine the way homicide offenders recall their homicide offence) and the procedure were

  verbally explained

  
  

  - Ps were asked to describe their homicide offences in as much detail as possible (open-ended interview) and

  prompted to provide details using Step-Wise Interview

  -interviewers were two senior psychology

  graduate students and one research assistant (blind to the PCL-R results).

  Post interview analysis

  - narratives were subsequently transcribed, as close to verbatim as possible.

  - Analysed using Wmatrix (used to compare parts of speech and to analyse semantic concepts) and DAL (used

  to examine the affective tone of the words).
• Results- instrumental worldview

  Psychopaths used more subordinating conjunctions i.e. 'so' and 'because' than non-psychopaths

  
  

  1.54% controls vs 1.82% of psychopaths.
• Results- Socioemotional needs

  Psychopaths used approx twice as many words related to psychological and safety needs, compared to the controls

  
  

  - controls used significantly more words relating to social needs.
• Results- Poverty of affect

  -Psychopaths used more past tense verbs and fewer present verbs than controls

  
  

  -Psychopathic language was significantly less fluent that the controls' language - 33% more disfluencies.
• What are the conclusions of Hancocks study?

  >Psychopaths more likely than non-psychopaths to describe cause and effect relationships when

  describing their murder

  
  

  > Psychopaths focus more on physiological needs (money/food/sex) than higher level social needs when

  compared to non-psychopaths

  
  

  > Psychopaths will linguistically frame their murders as more in the past and in more psychologically

  distant terms than non-psychopaths.
• What ethics were broken in Hancocks study ?

  >Mild deception - told that the aim was to examine the way homicide offenders recall their homicide offence hence did not mention psychopathy. unclear if they knew were being assessed for psychopathy or compared against others

  
  

  Informed consent - as above.

  
  

  >Debrief - Not debriefed regarding the purpose of the interview!

  
  

  >Right to withdraw - prisoners so despite being given right to withdraw they might not feel that this offer is genuine.
• Ethical guidelines adhered to

  >Confidentiality/privacy - no personal details included apart from age and gender

  
  

  >Degree of informed consent - Self-selected + they were given the aim with a small omission of psychopathy. They knew their language will be analysed

  
  

  Right to withdraw - given right to withdraw throughout.
• Similarity between Gould + Hancock

  both had quantitative data so both had objective, numerical data + therefore considered more scientific

  
  

  [E1] For example,

  Gould/Yerkes gathered IQ test scores to compare different groups' intelligence. For example, he found that

  average mental age of white American was 13

  
  

  [E2] Hancock gathered qualitative data (interview) but

  analysed it in a quantitative manner, e.g. psychopaths were found to use 33% more disfluencies in their

  language (such as erm, uh...) compared to controls.
• What is a difference between Gould + Hancock

  > sample size [E] which can affect population validity of the study

  
  

  [E1] Hancock had 52 participants in total, with only 14 male psychopath murderers. It would be difficult to

  generalise those results to all psychopaths - for example those who did not commit murder

  
  

  [E2] Gould

  reviewed Yerkes' study who used 1.75 million soldiers in his study, who will be more representative of male

  part of the population compared to Hancock.
• Applications of Hancock?

  Prison context with inmates? Problems with applying these findings in order to prevent and rehabilitate?

  
  

  Children in schools - early screening and intervention?
• How does Hancock link to individual differences area?

  [A] in order to understand complexity of human mind and behaviour we must focus on and investigate the differences between people rather than what we have in common

  
  

  [L] Hancock links to the area, because he is looking at differences between psychopaths and non-psychopaths, especially in the use of language

  
  

  [E] He found that psychopaths used their language differently to non-psychopaths, for example psychopaths were found to use 33% more disfluencies in their language (such as erm, uh...) compared to controls.
• Another way Hancock links to ID area

  [A] Another assumption suggests that individual differences can be measured through psychometric tests, such as personality inventories.

  
  

  [L] By training interviewees and conducting these interviews, we may have a different measure of psychopathy than the PCL-R in the DAL and Wmatrix

  
  

  [E] It was found that psychopaths used more subordinating conjunctions i.e. 'so' and 'because' than non-psychopaths (1.54% controls vs 1.82% of psychopaths).
• To what extent does Hancock change our understating of the key theme?

  understanding not changed very much as both psychometric tests have problems with validity

  
  

  > Yerkes designed a biased IQ test which grouped people as having differing intelligence levels by race when actually this test was based on cultural knowledge and Hancock classified his participants as psychopathic or not using a reduced threshold for the PCL-R (25 rather than 30)

  
  

  >Hancock found that psychopaths were more likely to be disfluent and to discuss their basic socioemotional needs than non-psychopaths. However, if some of the ps in the psychopath group were actually non-psychopaths, this would mean the results had questionable internal validity

  
  

  Yerkes found that black men were less intelligent than white men on average, but the questions were based on American culture or having attended formal schooling e.g. 'Crisco is a type of what?'. Therefore, to base someone's IQ score on a test which does not measure intelligence is also an invalid measure

  
  

  > Furthermore, both studies used quantitative data to measure differences. Yerkes quantified IQ as a mental age per racial group and Hancock used the computer programs, Wmatrix and DAL, to reduce the words used in the interview into numerical form and used this to compare between psychopaths and non-psychopaths.
• Ethnocentrism in Hancock

  • All from Canada - people in other countries could use language in different ways according to the nature of the language or culture (e.g. the extent to which physiological needs are a preoccupation in a particular culture)

  
  

  • Canada has two official languages - English and French - we have no idea whether they were speaking both so some disfluencies could be due to their first language being in fact French, whereas they were interviewed in English.
• Pop validity

  • All Canadian - limited generalisability to non-English speaking countries

  
  

  • All male - limited generalisability of results to females

  
  

  • All criminals - doesn't represent psychopaths who are not criminals

  
  

  • All convicted of the same crime (murder)

  
  

  • Volunteers - Will it represent other murdered who didn't volunteer?
• Reliability

  - High inter-rater reliability the PCL-R

  
  

  - interviews all followed same 'step-wise' interview procedure

  
  

  - DAL,WMATRIX used to analyse language- ensured each p approached in consistent way

  
  

  -interview - narratives varied in, for example, which parts of the crime they described in most detail

  
  

  ☹ 14 psychopaths not really large enough to establish a consistent effect
• Internal validity

  Aimed to reduce demand characteristics - did not mention psychopathy when explaining aims

  
  

  Data kept confidential, which potentially reduced socially desirable answers

  
  

  Double blind - reduced the likelihood of researcher bias during interviews

  
  

  Step-Wise - used to ensure that interviewers are well trained + they avoid leading questions. Valid measured used - Wmatrix, DAL

  
  

  > cut off points 25 normally 30 - were they psychopaths?

  
  

  - Language use- what about education could affect disfluencies
• Ecological validity

  - talked about their own crimes
• Qualitative vs quantitative data

  Collected qualitative (interviews) but analysed in quantitative manner

  
  

  Hancock could check whether differences in language use between psychopaths and non-ps were statistically significant.
• Controls

  • StepWise

  • DAL

  • Wmatrix

  • PCL-R