Yvonne Barnes-Holmes

1
Understanding the Clinical Processes in ACT

• Yvonne Barnes-Holmes
• Dermot Barnes-Holmes

2
Co-Authors

• Ian Stewart
• Louise McHugh
• Kelly Wilson
• Barbara Johnson
• Brandy Fink
• Andy Cochrane
• Anne Kehoe
• Hilary-Anne Healy
• Claire Keogh
• Jenny McMullen

• Carmen Luciano
• Francisco J. Molina Cobos
• Olga Gutiérrez
• Sonsoles Valdivia
• Marisa Páez
• Miguel Rodríguez
• Francisco Cabello
• Carmelo Visdómine
• José Ortega
• Francisco Montesinos
• Mónica Hernández
• Laura Sánchez

3
Introduction

• There is no theory behind therapy, the former is
  a coherent set of theoretical constructs that
  hang together and make predictions, the latter is
  a coherent set of techniques that make a
  different set of predictions
• Almost never in the history of psychology have
  they come together in a manner that was both
  theoretically consistent and technologically
  effective
• ACT is no different, but as the field develops,
  there is growing reason to believe that there is
  considerable overlap between Relational Frame
  Theory (RFT) and ACT and that the former can make
  sound predictions about why the latter works, and
  to some extent about what the latter should look
  like

4
Overview

• The current talk will review some of the
  predictions and empirical evidence that support
  processes and techniques identified in ACT
• For the sake of simplicity, and in order to be
  consistent with the evidence, we will divide ACT
  into the following
• Acceptance vs. Avoidance
• Acceptance vs. Cognitive Control
• Values
• Defusion

5
Acceptance vs. Avoidance
6
Acceptance vs. Avoidance

• Our first place to start looking at ACT (Study 1)
  was to analyse the distinction between acceptance
  and avoidance if this was not clear-cut, then
  the basic terminology might need to be
  reconsidered
• ACTs emphasis on the dichotomy between
  acceptance and avoidance and the development of
  the AAQ suggested that we might be able to
  functionally differentiate individuals in terms
  of their propensity towards acceptance or
  avoidance
• We took 15 undergraduates who were low in
  acceptance (at least 1 SD below the mean on the
  AAQ) and 14 high in acceptance (at least 1 SD
  above the mean)

7
Acceptance vs. Avoidance

• Participants were exposed to a simple automated
  task that required them to match nonsense
  syllables
• During the task, however, matching on some trials
  resulted in the presentation of a horrible
  aversive image (e.g. mutilated bodies) for 6
  seconds
• Participants were required to rate each aversive
  picture
• But, primarily we wanted to determine how long it
  took them to do the task when they had
  discriminated which type of picture would come
  next
• Our prediction was that low accepters/high
  avoiders would take longer to complete tasks,
  which they had learned would be followed by an
  aversive picture
• This, for us, was a type of avoidance

8
Median Reaction Times

• During the task, High Acceptance produced similar
  reaction times whether they expected to see
  either an aversive or a neutral image next, so
  anticipation or avoidance was limited
• But, Low Acceptance exhibited significantly
  longer reaction times when they expected to see
  an aversive image (p 0.015)

3
2.5
High
Low
2
1.5
1
0.5
0
A N A N
But could this be simply because the Low
Acceptance Group perceived the neutral pictures
to be more unpleasant and thus legitimtely more
avoidable than the High Acceptance group?
9
Self-Report Ratings

• No, because High Acceptance rated the aversive
  images as more unpleasant and more emotionally
  intense than Low Acceptance
• But yet, Low Acceptance were less willing to look
  at either images than High Acceptance

Pleasant Unpleasant

Mild Intense
Willing Unwilling
10
Discussion

• So, the outcomes were consistent with ACT
  predictions regarding acceptance and avoidance
  and their dichotomy
• Individuals low in acceptance/high in avoidance
  showed greater anticipatory avoidance of the
  negative pictures than those high in
  acceptance/low in avoidance
• This avoidance was consistent with their own
  ratings of willingness to look at the pictures
• Furthermore, this avoidance occurred even though
  these individuals rated the pictures as less
  unpleasant and less intense than the other group
• The high acceptance groups, therefore, showed
  less avoidance and greater experiential
  willingness in the face of adversity outcomes
  that are consistent with ACT predictions

11
ERPs and Avoidance

• Study 2 replicated Study 1, but incorporated
  Event Related Potentials (ERPs) during the task
  with
• 6 High Acceptance
• 6 Low Acceptance
• 6 Mid-Range Acceptance
• Once again, we predicted that level of avoidance
  would differentiate and we hoped it would be
  detected by the ERPs

12
Median Reaction Times

• Identical to Study 1, High and Mid Acceptance
  produced similar reaction times for both aversive
  and neutral images, showing no anticipation or
  avoidance
• But, Low Acceptance again emitted longer reaction
  times when they expected to see an aversive,
  rather than a neutral, image (p 0.0431)

3.5
3
2.5
2
Low
1.5
High
Mid
1
0.5
0
A N A N A N
13
Self-Report Ratings

• Again, this was not because the pictures were
  less unpleasant, because the High and Mid
  Acceptance rated the aversive images as more
  unpleasant and emotionally intense than Low
  Acceptance
• But, Low and Mid Acceptance were less willing to
  look at the images

14
ERPs Recordings

• As expected, the ERPs recordings discriminated
  between the two types of pictures, with the
  unpleasant pictures producing significantly more
  positive wave forms than the neutral pictures for
  all groups

• And an interesting finding emerged with regard to
  the scalp locations . . .

15
Low Acceptance
Area Dimensions (?V ms)
High Acceptance
16
ERPs Recordings

• The fact that the Low Acceptance group showed
  greater negative activation for left hemisphere
  electrodes could suggest greater verbal activity
  for this group, which might indicate the use of
  verbal avoidance strategies (e.g. This is not
  real, think of something else, etc.)

17
Discussion

• So again, the avoidance groups could be
  distinguished from one another on several
  predictable counts -- Low Acceptance showed
  greater anticipation of the aversive images than
  the others and were less willing to look at them
  -- and yet, they rated the pictures as less
  unpleasant
• Some willingness distinctions even emerged
  between mid and high range accepters
• The unwillingness and tolerance avoidance for Low
  Acceptance was associated with greater negative
  activation for left hemisphere electrodes,
  suggesting the activation of verbal areas
• Again, the former outcomes are consistent with
  ACTs emphasis on acceptance, avoidance and
  willingness and the ERPs data were consistent
  with RFTs emphasis on verbal behaviour

18
Acceptance vs. Cognitive Control
19
Acceptance

• Up until the mid-90s, CBT was still insistent
  that explicit attempts to control cognitive
  events directly would reduce their frequency and
  impact, and thus be associated with positive
  clinical outcomes
• ACT has always offered a counter-approach because
  of its contextualistic underpinnings that argues
  that the only way to change verbal events is to
  change the context in which they occur and
  acceptance is the term we use to describe this
  broader target
• In this regard, though not intentionally, ACT is
  more in line with Eastern traditions that
  emphasise acceptance/mindfulness
• But Eastern traditions are not sciences and thus
  cannot be relied upon to provide scientific
  argument or evidence

20
Acceptance

• Although in Eastern traditions and in ACT, we had
  reason to believe that acceptance was an active
  ingredient in positive clinical outcomes and
  psychological well-being generally, there was
  almost no empirical evidence to attest to this
• Furthermore, positive empirical evidence for the
  impact of acceptance would to some extent
  undermine positivity for the main existing
  alternative that was cognitive control which
  functionally may be seen as the opposite of
  acceptance
• It should also be added that empirical evidence
  for cognitive control as an active ingredient in
  CBT is relatively scarce, in spite of its wide
  usage

21
Acceptance

• So, thus far, we had some comfort in the
  terminology that suggested a dichotomy between
  acceptance and avoidance
• But, acceptance as a clinical tool was something
  else
• In our first empirical analysis of acceptance as
  a mechanism of change, we set out with a very
  simple aim -- to see if we could construct a
  short, but potent, acceptance intervention that
  would be functionally similar to what is
  presented in therapy, but which might just work
  in an experimental context
• This was demonstration research of the simplest
  kind

22
Study 3

• During Study 3, normal participants were simply
  presented with a computerised task in which they
  were asked to match a lot of neutral pictures and
  a small number of horrible aversive pictures
  (e.g. mutilated bodies)
• The former pictures simply represented an
  experimental control, while the latter
  represented our core effort to provide
  participants with a clinical strategy they could
  use to deal with unpleasant psychological/visual
  content

23
Avoiding Negative Images

• Because the matching was too simple to function
  as a dependent variable, we targeted
  participants willingness to look at the aversive
  pictures by (1) giving them the option to avoid
  the pictures altogether before the trial and
  counting how many they looked at and (2)
  observing how long they would endure them on
  screen

24
Acceptance or Control

• Participants were exposed to the baseline
  matching task, the intervention, and then the
  task again
• Both interventions involved the presentation of a
  vignette in which participants were asked to --
  imagine that they had witnessed a horrific car
  accident in which they had to rescue the badly
  injured and bloodied victims from the car and to
  imagine that they found the sight of blood
  extremely aversive
• They were then given a coping strategy/interventio
  n to help them deal with the vignette (and to
  influence their subsequent performances on the
  negative pictures)

25
Acceptance vs. Control

• Participants in Cognitive Control were instructed
  to try to control their emotional reactions and
  to avoid feelings of discomfort (e.g. by imaging
  that the blood was just like tomato ketchup)
• Participants in Acceptance were instructed to
  fully embrace their feelings of discomfort (i.e.
  to fully accept that trying to save the bloodied
  and mutilated victims would be the most horrific
  experience of their lives)

26
Experimenter Influence

• Experimenter influence were also manipulated by
  altering the instructions and the extent to which
  the experimenter monitored the matching
  performances
• During the No Instruction/No Monitoring
  conditions, participants were informed that it
  did not matter whether they looked at the
  negative pictures (i.e. no instruction) and the
  experimenter sat approximately 30 feet away and
  pretended to read a book (no monitoring)
• During the Instruction/Monitoring conditions,
  participants were told that it was very important
  to look at the negative pictures (instruction)
  and the experimenter walked around actively
  monitoring performances (monitoring)

27
Results

• The results of the study failed to differentiate
  between the two groups on the number of aversives
  observed
• However, they did differ in their mean response
  latencies while the aversives were on the screen
  (i.e. aversive tolerance time)

28
Mean Response Times Neutral Pictures








On the neutral pictures, there were no changes at
all between Baseline and Post-intervention, as
expected
29
Mean Response Times Aversive Pictures

• But, on the aversive pictures, Acceptance and
  Control differed significantly when combined with
  Instruction/Monitoring (p 0.002)
• Strategy and Experimenter Influence interacted
  significantly

30
Discussion

• The Acceptance strategy increased participants
  tolerance time in the presence of the aversive
  pictures (when combined with active experimenter
  influence)
• Control did not and decreased tolerance in both
  cases
• While both strategy outcomes appeared to be
  influenced by the social context, further
  analyses indicated that this primarily affected
  the extent to which participants applied the
  strategies, rather than affecting the strategies
  directly (i.e. the strategies were applied more
  when the experimenter attended)
• This was our first empirical evidence that
  acceptance could be delivered as a brief
  therapeutic intervention in an experimental
  context and was associated with positive outcomes
• Cognitive control was in fact counter-productive
  in terms of altering aversive tolerance when the
  images were present

31
Acceptance vs. Control with Pain

• In Study 4, we were concerned that the data so
  far would not generalise to physical pain and the
  psychological content associated with that
  perhaps different outcomes would emerge relative
  to coping with aversive visual imagery
• So, we exposed participants to systematic
  electric shocks
• This was based on a previous study by Gutierrez,
  Luciano, Rodriguez, and Fink who compared
  acceptance and control as coping interventions
  with electric shock with 40 undergraduates
• They reported that Acceptance not only increased
  shock tolerance, but also reduced participants
  believability of their own subjective pain ratings

32
Our Study

• Although the original study was entirely
  consistent with our own findings thus far, there
  was increasing concern within the community about
  experimental precision but this was hard to
  offset against external validity
• So in Study 4, we tried to come up with a format
  that was fully automated (hence experimentally
  clean), but that would still allow the
  interventions to be impactful
• We did some refinement of the Acceptance and
  Control exercises and metaphors to remove
  possible confounds
• And we began to look at values as an active
  addition to acceptance

33
Design

• 40 normal participants were assigned to four
  conditions

Intervention Values Context Pre-Intervention Post-Intervention
Acceptance High
Acceptance Low
Control High
Control Low
34
Delivery

• The entire procedure was automated through a
  program containing a series of video clips
• Participants progressed through the clips at
  their own pace, individually and alone
• Clips were rated first by independent observers,
  for consistency, adherence and empathy and were
  found to not differ in any capacity

35
Delivery
36
Acceptance High/Low Values
A Participants were provided with metaphors and
experiential exercises indicating that the best
way to deal with pain related thoughts and
feelings was to accept them in the context of
whatever action is being taken
HV Participants were asked to imagine that they
suffered from chronic pain and that the task
involving shock was one which they must do in
order to support their family LV Participants
were told that the aim of the experiment was to
contribute to research on the relation between
voltage level and perception of shock
37
Control High/Low Values
C Participants were given metaphors and
experiential exercises designed to teach them
that the best way to deal with pain related
thoughts and feelings was to distract themselves
by imagining pleasant images
HV Participants were asked to imagine that they
suffer from chronic pain and that the task
involving shock was one which they must do in
order to support their family LV Participants
were told that the aim of the experiment was to
contribute to research on the relation between
voltage level and perception of shock
38
Shock Tolerance Data

• The Acceptance participants significantly
  increased their shock tolerance from pre- to
  post-intervention
• Control produced no change

39
Self-Report Data
Low Pain

• There was an interesting effect for values
  although there was no significant main effect,
  High Values participants rated the pain as
  greater across time, whereas Low Values rated it
  as less

40
Tolerating High Pain

• We wanted to check whether some of the effects
  were driven by people who had different
  perceptions of how much pain they were in -- so
  we examined only those reporting great pain more
  closely
• 100 of participants in Acceptance who reported
  greater experienced pain Post-Intervention showed
  an increase in tolerance levels, compared to only
  50 of the same sub-set of Control (significant
  p 0.0455)
• We also analysed the number of trials for which
  participants continued in the Post-Intervention
  task after reporting high levels of pain (gt 80)
  and found that the median number of trials for
  Acceptance was 4, compared to 2 for Control
  (significant p 0.0069)

41
Discussion

• So as an intervention, Acceptance worked better
  than Control in the context of experimentally
  physical pain in the form of electric shock
• Changes in tolerance were particularly strong for
  participants experiencing a lot of pain and using
  Acceptance
• The effects were the same as those reported by
  other researchers even in a highly structured
  automated experimental environment
• While the Values manipulation did not have a
  significant effect on shock tolerance, it did
  affect self-reports of pain, in that participants
  in High Values reported more pain subsequent to
  the intervention (perhaps the values component
  oriented them more towards their pain, but not in
  an avoidant way)

42
Study 5 Simple Rules

• One issue that had been emerging across
  experiments was the possibility that participants
  were not really engaging with the various
  features of the interventions (i.e. the exercises
  and metaphors), but that they were simply
  generating or following simple rules
• So, in Study 5, we compared the full Acceptance
  and Control interventions used before, but added
  two new interventions that simply comprised of an
  Acceptance Rule and a Control Rule -- a brief and
  simple rule for accepting or distracting
• In this study, we also employed a Placebo
  Condition

43
Experimental Conditions
Acceptance Rule Acceptance Rule, Metaphor Exercise Control Rule Control Rule, Metaphor Exercise Placebo
44
Tolerance Data

• Only Full Acceptance increased tolerance
  significantly from Pre- to Post-Intervention, but
  none of the other four
• Distraction-Rule actually decreased tolerance
  significantly

45
More Pain More Shocks

• Again, we looked at those participants who
  reported more pain and still took more shocks and
  found that these were mostly in the Acceptance
  Conditions

46
Discussion

• So, the positive acceptance outcomes thus far
  could not be explained in terms of simple rule
  following the metaphors and exercises were
  essential
• When these were absent, the moderate improvement
  in pain tolerance for an acceptance rule was
  non-significant
• Although Distraction effects are again negligible
• Distraction actually makes you worse when it
  comes in the form of a simple rule

47
Different Pain Same Outcome

• The next study (Study 6) was also concerned
  acceptance, but attempted to broaden the
  generality of the work by employing a new type of
  pain induction, that might circumvent criticisms
  that electric shock is not a good analogue of
  clinical pain
• So, three groups of participants were assigned
  to
• Acceptance
• Control
• Placebo
• And were exposed to the radiant heat pad in a
  fully automated procedure

48
Heat Apparatus
49
Results

• At baseline, the groups did not differ on a
  series of psychological measures
• And the amount of heat tolerance was tightly
  controlled

50
Tolerance Data

• Both Acceptance and Control increased pain
  tolerance, but only Acceptance was significant

51
Discussion

• So, positive outcomes again for acceptance now
  a total of six experiments
• Acceptance is always significantly better than
  Control, which had negligible effects
• Outcomes so far have included tightly controlled
  experimental environments, a range of populations
  and numerous experimental methodologies and types
  of pain
• The data overall are highly consistent with ACTs
  centrality for acceptance and its predictions on
  avoidance
• The ERPs data were consistent with both ACT and
  RFT and added legitimacy to the outcomes and
  methodologies

52
One Query?

• But one thing troubled us and we had seen it in
  research by other labs
• In some studies, there had been positive (albeit
  limited and never significant) outcomes for
  Cognitive Control
• So, in the radiant heat research, we began to
  look more closely at our interventions and those
  used in other studies
• In the heat study, in particular, we noticed that
  part of the Control intervention involved saying
  a pain-related thought aloud before participants
  tried to distract themselves from it

53
Revisions

• So, we thought that it might just be possible
  that this feature offered a type of defusion, or
  at least cognitive distancing, that may have
  attributed to the outcomes
• And we set about modifying the Control
  intervention so as to eliminate this potential
  confound (Study 7)
• Our new condition was called Control Revised
• And we were amazed at what we found . . .

54
Tolerance Data

• The effects for Acceptance were exactly the same
• But, Control had no effect at all, and in fact
  increased pain tolerance was decreasing

55
Discussion

• So, even the small improvements that had been
  previously recorded for Cognitive Control may not
  have functioned in the way that was intended
• Some of the experimental interventions had
  spurious features that enabled aspects of
  defusion to creep into the Control protocols
• In our latter heat experiment in which this
  feature was addressed directly, the effects for
  Control could not be differentiated from Placebo

56
Values
57
Investigating Values

• But, of course, there is more to ACT than
  acceptance and much of what we do in the therapy
  depends upon the combination of active
  ingredients rather than simply a series of
  incoherent or unintegrated steps
• However, as much as possible, we try to isolate
  the components individually for experimental
  purposes to get a better understanding of
  outcomes and processes
• So, we turned our attention next to Values
• But note, that where we had looked at values
  before, the outcomes were mixed and it would be
  very difficult to deliver values as a solitary
  intervention

58
Values

• We have done only one study (Study 8) to date
  looking specifically at values
• This study was conducted in Spain and attempted
  primarily to assess the influence of a values
  clarification exercise
• Although two types of exposure to painful private
  events were also compared (writing down versus
  experiential exercise) across three conditions

Values Clarification Values Clarification Writing Values Clarification Experiential Exercise
59
Values

• 10 participants were assessed on personal
  barriers, valuable actions and areas of valued
  living affected by problems and barriers

Subject 2
Values Clarification

• Values Clarification alone quickly and steadily
  reduced barriers and improved reports of valued
  living and effect enhanced across time

60
Values Clarification Writing

• Values Clarification Writing alone showed a
  similar outcome, but the decrease in barriers was
  less
• Values Clarification Exercise alone was similar
• Overall, the type of exposure to private events
  did not matter greatly, and these even softened
  the effects relative to Values Clarification
  alone

Values Clarification Exercise
61
Discussion

• So, some positive effects for values
  clarification
• No matter, how you do it, a simple values
  clarification exercise helps to increase the
  extent of actual valued living and decrease
  barriers to same
• There were some minor differences in terms of how
  this can be done, but these were minimal
• The data also identified what appeared to be a
  functional relationship between decreases in
  barriers and improvements in valued living
• These are entirely consistent with ACT
  predictions regarding how private events can
  function as barriers and how these can be altered
  with values

62
Defusion
63
Defusion

• But, no-one would think for a second that ACT
  would be ACT without defusion
• In fact, defusion, it seems is the gel that glues
  the active ingredients together
• In fact, acceptance is often difficult when
  defusion is not in place
• Also, for RFT the deliteralisation effects that
  underpin defusion techniques are central to ACTs
  outcomes, so in ways studying defusion is perhaps
  the best test of the relationship between the
  theory and the therapy

64
Defusion

• When we started looking at defusion, we had only
  one previous study by Masuda et al. (2004) to
  work from
• They attempted to assess the impact of word
  repetition on believability and discomfort levels
  associated with negative self-relevant words
  (e.g. anxious, anxious, anxious etc.)
• Their findings indicated that the use of a
  defusion rationale produced greater reductions in
  discomfort and believability about the words when
  compared to a thought suppression rationale or a
  distraction task

65
Defusion

• In this study (Study 9) , we automated the
  presentation of 20 positive and 20 negative
  self-statements
• This generated a total of 60 statements because
  there were three exposures to each statement
• After the appearance in screen of each statement,
  participants were asked to provide ratings
  regarding their reactions to the statements in
  terms of
• Comfort
• Believability
• Willingness

66
Defusion

• We manipulated defusion in two ways
• (1) Defusion Instructions
• The 80 undergraduates were randomly assigned to
• Defusion Condition (pro-defusion instructions)
• Anti-Defusion Condition (anti-defusion
  instructions)
• Neutral Condition (neutral-defusion instructions)

67
Defusion Instructions

• In the current experiment, we are interested in
  the emotional impact of unusual self-statements.
  The scientific literature in this area shows that
  if you rephrase a self-statement like I am an
  awful person into I am having the thought that
  I am an awful person, then the emotional impact
  of the statement is reduced
• In other words, thinking or saying words like I
  am having the thought that I am an awful person
  is easier to deal with than simply thinking or
  saying I am an awful person

68
Defusion

• (2) Defusion in Visual Format
• We wanted to see the extent to which defusion
  within the visual presentation of the
  self-statements would give rise to
  defusion-predictable outcomes
• To manipulate this, we employed three types of
  presentation format for each statement
• Normal
• Defusion
• Abnormal

69
Normal Negative Self-Statement
Deep down there is something wrong with me
70
Defusion Negative Self-Statement
I am having the thought that deep down there is
something wrong with me
71
Abnormal Negative Self-Statement
I have a wooden chair and deep down there is
something wrong with me
72
Results Comfort

• The (pro) defusion instructions were correlated
  with less discomfort than the other two types of
  instruction
• As was the defusion presentation format

73
Results Willingness

• The (pro) defusion instructions were correlated
  with more willingness than the other two types of
  instruction
• As was the defusion presentation format very
  similar results to comfort ratings

74
Results Believability

• Contrary to predictions, the (pro) defusion
  instructions were correlated with more
  believability than the other two types of
  instruction
• As was the defusion presentation format very
  similar results to comfort and willingness ratings

75
Discussion

• Although they looked impactful in the ratings,
  the defusion instructions did not have a
  significant influence
• However, the Defused presentation format
  significantly decreased discomfort, increased
  willingness, but unexpectedly increased
  believability
• However, on closer inspection of the data and
  other information gathered from participants it
  may be the case that they were rating the
  believability of whole statements I am having
  the thought that . . rather than the content
  itself this is not unlike defusion

76
Discussion

• So, increases in willingness to having negative
  self-referential content were consistent with
  ACTs predictions regarding defusion
• Believability ratings, upon closer inspection,
  suggested that the defused format decreased
  participants believability of the content
  directly
• Decreases in discomfort were not directly
  predicted by ACT, but such outcomes are positive
  although they would not be targeted directly

77
Defusion Interventions

• In the previous study, we had assessed simple
  impacts for defusion and found that it generated
  positive and largely ACT consistent outcomes even
  when defusion occurred within the visual
  presentation of the content
• But, if we employed defusion as an intervention,
  as had been the case for Masuda et al., would we
  find similar outcomes?
• Study 10 attempted to address this question

78
Study 10

• Participants generated a personalised negative
  self-relevant thought that represented a summary
  of several related personal statements
• They were then given a written protocol that
  contained an instruction followed by an exercise
• The three protocols were
• Defusion
• Thought Control
• Placebo

79
Experimental Conditions
Rationale Exercise
Defusion Defusion
Thought Control Thought Control
Defusion Thought Control
Thought Control Defusion
Defusion Placebo
Thought Control Placebo
Placebo Defusion
Placebo Thought Control
Placebo Placebo
80
Method

• Once again, the emotional impact of the negative
  self-referential statements was measured in terms
  of
• Discomfort
• Believability
• Willingness

81
Results Comfort
Uncomfortable
100
80
60
40
20
Pre-Intervention
0
Post-Intervention
Comfortable
Condition

• All interventions with a defusion component
  generated decreases in discomfort
• But, the largest effects were DD and PD,
  suggesting activity in the defusion exercise

82
Results Comfort

• Interestingly, the only significant differences
  pre- and post-intervention emerged for the
  following conditions
• Placebo-Defusion
• Defusion-Placebo
• Defusion-Defusion
• Thought Control-Thought Control

83
Results Believability
Believable
Pre-Intervention
Post-Intervention
Unbelievable
Condition

• All effects were in the right direction of
  decreasing believability
• But, D-D and TC-D showed largest decreases in
  believability

84
Results Believability

• The only significant differences pre- and
  post-intervention emerged for the following
  conditions
• Placebo-Defusion
• Placebo-Thought Control
• Defusion-Placebo
• Defusion-Defusion
• Defusion-Thought Control
• Thought Control-Defusion
• Thought Control-Thought Control
• So, a very mixed bag overall

85
Results Willingness

• All effects were in the right direction of
  decreasing unwillingess
• But, D-TC was the only significant outcome

86
Discussion

• Quite a mixed bag overall
• But, generally most positive effects in predicted
  directions for packages containing defusion
  features
• Defusion exercise appeared to be somewhat more
  effective than a simple rationale

87
Concluding Comments
88
Concluding Comments

• There are many more analogue studies completed
  and underway than those reported here
• The effects for ACT components across the board
  are predominantly as predicted and compare
  favourably with substantively weaker outcomes
  generated by target comparisons
• The range of issues generated by the studies
  shows the complexity of the effects and the
  difficulty in conducting high quality research in
  this modality
• As studies progress, the standard of experimental
  rigour is exceptional

89
Concluding Comments

• Automated interventions
• Balancing for gender
• Balancing for heat tolerance, acceptance etc.
• Pre-screening with relevant psychological
  assessments
• Including self-report measures
• Blind experimenter
• Use of different types of physical and
  psychological stressors
• Use of non-clinical populations
• Very substantive N in some cases
• Interventions are very closely matched,
  topographically and functionally
• Range of ACT components tested

90
Concluding Comments

• We are now in a place where these types of
  analyses can be done effectively and with high
  levels of precision
• The evidence is overwhelmingly positive . . .