BENDERGESTALT II

1
BENDER-GESTALT II

• Scott L. Decker, Ph.D.
• Georgia State University

2
Overview of presentation

• History of Bender-Gestalt
• Previous Research
• BG II Revision Goals
• BG II Features and Psychometrics
• Administration/Scoring
• Interpretation and Derived Scores

3
Brief Background
Georgia State University
4
Bender-Gestalt

• By Lauretta Bender, 1938
• One of the most used psychological tests.
• Developmental test for children and
  psychopathology in adults
• Individually administered, paper and pencil test
  that contains 9 geometric figures.

5
Bender-Gestalt

• Original figures from Wertheimers (1923) research
  in perception
• Pragnantz and the laws of perception

6
L. Bender

• Bender (1938) operated under the assumption that
  the visual gestalt function is a fundamental
  function associated with language ability and
  closely associated with various functions of
  intelligence such as visual perception, manual
  motor ability, memory, temporal and spatial
  concepts and organization (p. 112).

7
Constructs and Models

• Visual-Motor Perception and Integration
• Bender Maturation and psychopathology
• Hutt psychodynamic projective personality
• Koppitz Development
• Lacks Neuropsychological screener

8
Koppitz Model
9
Overview Previous Research

• Piotrowski (1995)
• mainstay in the assessment batteryas an
  assessment tool in appraisal of intelligenceas a
  screening technique for neuropsychological
  dysfunction, a clinical tool for sampling
  visual-motor proficiency, and as a standard
  projective technique in the assessment of
  personality (p.1272)

10
Historic Uses

• Established Relationship
• Achievement
• IQ
• Brain Damage

• Not Established
• Localized organicity
• Personality/psychopathology
• Differential Diagnosis

11
Neuropsychology

• Construction
• Lacks (1999)

12
Visual Pathways
13
Neuropsychology
14
Visual Disorders

• Agnosia inability to visually identify
• Hemianopia half loss of visual field
• Scotoma small lesions, spots in visual field
• Optic Ataxia deficit in visually guided hand
  movements
• Prosopagnosia facial agnosia
• Alexia inability to read
• Neglect

15
(No Transcript)
16
Scoring Systems (Bender)

• evaluated the overall quality of each design on a
  scale that ranged from 1 to 5 on one design to 1
  to 7 on others.
• provided detailed descriptions of each point on
  the scale for each design.
• advocated the use of global scoring systems over
  error based systems, which she felt
  oversimplified the processes involved and failed
  to do justice to the test.

17
Scoring System by Keogh and Smith in 1961

• involved rating each design on a five-point scale
  based on the overall quality of the production.
  A score of 1 would be given when a figure was
  unrecognizable and a score of 5 would be given
  when all parts of the figure were present and
  recognizable. Although the system generated some
  research, the authors never provided normative
  data.

18
deHirsch, Jansky, and Langford (1966)

• The scoring system used a single global
  inspection procedure that yielded a pass or fail
  for each design. The system evaluated the
  essentials of the gestalt and the degree of
  differentiation of each design.

19
Brannigan and Brunner (1989, 1996, 2002)

• six designs (A, 1, 2, 4, 6 and 8) as deHirsch,
  Jansky, and Langford (1966) and Jansky and
  deHirsch (1972), a six-point scoring system was
  devised for greater differentiation in scoring
  each design. Scoring ranged from 0, for random
  drawing, scribbling, having no concept of the
  design, to 3, all major elements present and
  recognizable with only minor distortions, to 5,
  accurate representation.

20
Error Based

• Koppitz (scoring by statistical fiat, classroom
  of children, teacher opinion)
• Lacks
• Hutt Pascall
• Mixed construct validity (rotation distortion
  simplification ?)
• Does not capture whole design (see ASB,
  Innovative Features of BG II (Brannigan, Decker,
  Madsen)
• Similar to pathognomic signs

21
History of Scoring

• Qualitative methods based on rating scale
  demonstrate high reliability and validity,
  (Sattler)
• Capture whole performance
• Clear construct dimension

22
Overview of new Developments
The revision of the BGT is currently taking
place, and includes 1. New items 2. Memory
recall 3. National norms (N5,000) 4. Clinical
validity 5. Time and planning estimates 6.
Quantitative/Qualitative scoring 7. Co-normed
with Stanford-Binet 8. Test observation form
23
Item Selection

• Item Generation
• Expert Rating Scale (Difficulty, Inclusion)
• Item Tryout
• Scoring Criteria Analysis
• Selection of New Items

24
Some Difficulties

• Measuring pragnantz?
• How to select gestalt items.

25
Initial selection of items We began by
generating a large number of possible items.
Some of these items were created to lower the
floor of the test (very easy items) and some to
raise the ceiling (very difficult items). The
initial set of 66 items were reviewed by our
panel in order to select the most promising items
for further testing.
26
Rating
Scale Reviewers were asked to rate each item in
terms how difficult they felt it was (Very Easy
to Very Difficult on a 4 point Likert scale),
and desirability/compatibility with existing
designs (Definitely Exclude to Definitely
Include on a 4 point Likert scale.)
27
Rating scale analysis Responses to the
difficulty and inclusion scales were analyzed
separately with WINSTEPS in a two stage process.
The inclusion scale was used to select the items
which raters were more likely to approve (high
probability of Include or Definitely
Include). The mean person measure of -0.33
indicates that our raters were more likely to
exclude items than to include them, however due
to the RMSE of .38 for item measures we selected
all items with an inclusion measure below 0
(lower measures mean the item is easier to
include).
28
Rating scale analysis The second stage in
the selection process involved the perception of
item difficulty. Because we are interested in
items for the top and bottom of the scale, we
used the ratings of item difficulty to select
from the high inclusion items. Items which were
rated very easy or very hard were selected for
the testing by preschoolers and
adolescents/adults, respectively. We therefore
cut down the number of items we needed to test
from 66 to 16 (6 easy and 10 hard) based on the
initial review process.
29
Scoring the original 9
designs In order to compare the difficulty of
the new designs with those in the original
Bender-Gestalt Test, we first had to arrive at a
method for scoring those items. The original
Bender-Gestalt test had been administered with
the Stanford-Binet tryout study in 2000.
Responses from a sample of examinees between the
ages of 5 and 13 were selected and scored using a
rubric designed to measure the accuracy of
responses. This initial rubric specified 3
criteria to check for each item, yielding a score
of 0-3 points for each item.
30
Item Tryout
Two forms were created to test the new
items. The preschool form contained the 3
easiest items from the current Bender-Gestalt
Test, as well as the 6 new items. The
adolescent/adult form contained all 9 items from
the current BGT, as well as 10 new items. For
the purpose of comparing performance on the new
items with that on the original 9 items, earlier
cases which had only taken the 9 BGT items were
analyzed with the new sample.
31
Scoring Systems
Global Scoring rating scale Error Based -
Certain errors associated with different
pathologies. Accuracy Based (criteria of how
close the reproduction is to the actual
figure) Qualitative signs from test observation
32
Scoring Multimethod
Scoring criteria from a variety of methods
(from the 1930s to the present) were studied.
These criteria often include a variety of errors
within a single criteria. For example, either of
the reproductions below would qualify for
Simplification in the Lacks system because the
complete design has been broken into its
component parts.
33
Scoring
Identification of Errors Criteria combined,
redundancies removed Can be recombined into the
original scoring system at a later date.
Characteristics of these errors can be
examined, and their relationship to different
diagnoses tested. When the various methods and
criteria are combined, a total of 436 errors are
possible for the 16 designs on the BGT-R.
34
Examples of Errors

Test Item
Rotation error
35
Examples of BGT-R Errors
Retrogression error
Test Item
36
Examples of BGT-R
Errors
Simplification
Test Item
37
Scoring Global Scoring System

• 0 Random
• 1 Slight Vague Resemblance
• 2 Some Moderate Resemblance
• 3 Strong Accurate Resemblance
• 4 Nearly Perfect

38
Example of items rotation - 30 degree rotation
- 45 degree rotation - 90 degree rotation -
mirror image Rotation - one figure rotated
relative to other Closure Difficulty - Shapes
fail to touch Closure Difficulty - Shapes
overlap Closure Difficulty - Sketching or
reworking at point of intersect Closure
Difficulty - Distortion of the figure at point of
intersect Perseveration - Drew figure multiple
times Simplification - Shapes greater than 1/8
inch apart Simplification - Diamond as
disconnected pieces Distortion - missing or
extra angle(s) in diamond Distortion - either
shape misshapen Distortion - double
line Retrogression - square or triangle for
diamond Fragmentation - Part of design
missing Fragmentation - Drew only left half of
figure Fragmentation - Drew only right half of
figure
39
Accuracy

• ____All dots are round and solid. Column 1
  contains 1 dot column 2 3 column3 5 column4
  7.
• ___Alignment Middle dots 1, 5, 8, and 10 (see
  picture) can be connected with a straight line
• ___Left to right downward slant of dots above
  middle dots (1, 5, 8, and 10). Left to right
  upward slant of dots below middle dots (1, 5, 8,
  and 10)

40
Scoring

• All systems correlated
• Global Scoring System
• High reliability
• High correlation with Accuracy
• Easy to use, understandable
• Sensitive to clinical groups

41
Administration

• Copy and Recall
• Test Observation Form
• Motor and Perception

42
Administration Procedure

• Ages
• Start Item
• End Item
• 4 years through 7 years, 11 months
• Items 1-13
• 8 years and above
• Items 5 -16 

43
Observation Form
44
Administration
45
Motor Perception Tests
46
Demonstration

• Find a partner
• Get several pieces of blank paper
• Pencil
• Read Directions

47
Administration Copy

• I have a number of cards here. Each card has a
  different drawing on it. I will show you the
  cards one at a time.
• Use this pencil (give pencil to examinee) to
  copy the drawing from each card onto this sheet
  of paper (point to the drawing paper). Try to
  make your drawings look just like the drawings on
  the cards. There are no time limits, so take as
  much time as you need.
• Do you have any questions? Here is the first
  card.

48
Administration Recall

• Now I want you to draw as many of the designs
  that I just showed you as you can remember. Draw
  them on this new sheet of paper. Try to make your
  drawings just like the ones on the cards that you
  saw earlier. There are no time limits, so take as
  much time as you need.

49
Switch Partners

• Administer Copy
• Administer Recall

50
Example of BGT-II performance.
51
Scoring Global Scoring System

• 0 Random
• 1 Slight Vague Resemblance
• 2 Some Moderate Resemblance
• 3 Strong Accurate Resemblance
• 4 Nearly Perfect
• (pdf file)

52
Standard Scores

• Global Scoring System
• Norms table (pdf 1)
• Copy Recall (no 4 year old recall)
• Labels

53
Standardization

• N 4,000
• Age 4-85
• Co-normed with Stanford-Binet Fifth Edition
• Age, Sex, Race, Geographic Region, Education (p.
  33)

54
Descriptive labels for Bender-Gestalt II standard
scores 

• 145-160  Extremely advanced
• 130-144.1  Very high or very advanced
• 120-129.1  High or advanced
• 110-119 High average
• 90-109 Average
• 80-89 Low average
• 70-79 Low or borderline delayed
• 55-69 Very low or mildly delayed
• 40-54 Extremely low /delayed

55
Psychometrics

• Reliability .91
• Test-Retest .80
• Inter-rater .80-.90
• Concurrent
• Factor Analysis
• Growth (pdf)
• Clinical Studies

56
Growth Curve (p. 59)

• Rise and Fall of Visual-motor ability
• Increases sharply at young ages
• Levels off around age 16 to 49
• Declines around age 50
• Normative base allows for better geriatric
  assessments

57
Validity

• VMI, Koppitz
• Achievement
• WIAT, WJ III
• Cognitive
• WISC, WAIS

58
Factor Analysis WISC
59
Clinical Validity

• Matched Sample
• Non-clinical subject from normative sample
• Matched on Age, Sex, Race, Parent Education
• For multiple matches, random number used
• Statistical significance and Effect Size used.
• General Effect Size is Large (Cohen, 1987 .2
  small, .5medium, .8 gt large).

60
Clinical Validity Groups (p. 61)

• Gifted
• Mental Retardation
• Developmental Delay
• LD (Reading, Math, Spelling)
• ADHD
• Emotional Disorders
• Autism

• Alzheimers

61
Interpretation

• ..the need is not so much for more tests as it
  is for a better and more complete utilization of
  existing tests. Koppitz (1963)
• Multidimensional interpretation
• time and energy..curtailed..rarely
  unproductive..

62
Interpretation

• visual problems
• physiological limitations associated with
  illness, injury, fatigue,
• or muscular weakness
• physically disabling conditions, such as low
  birth weight,
• cerebral palsy, or sickle cell anemia
• environmental stresses
• impulsiveness
• inadequate motivation
• emotional problems
• mental retardation
• social or cultural deprivation

63
Additional Factors

• Maturation
• Attention
• Memory
• Personality
• Attitude

• Skill
• Cultural
• Motivation
• Psychopathology
• Executive skills

64
Interpretation

• taking an excessively long time to complete the
  task
• tracing the design with a finger before drawing
  it
• anchoring by placing a finger on each portion
  of a design
• while drawing it
• glancing briefly at a design and then drawing it
  from memory
• turning the design card or drawing paper in
  order to complete
• the drawing

65
Clinical Cases WNL
66
Assessment Process

• Review Referral Info
• Obtain Background, Social, and Setting
  Information
• Hypothesis Development
• Select Administer Tests

• Interpret results
• Intervention Strategies/Recommendation
• Report
• Communicate with Teacher/Parent
• Follow up

67
Decision Making Context

• Review Background/Referral Info
• Develop Hypotheses/Prediction
• Attention, Simple Vision, Perceptual, Motor,
  Motivation, Learning Disability
• Test
• Confirmation from multiple sources
• Revise Hypotheses

68
Case Study

• Case 1
• Case 2
• Case 3
• Case 4

69
Review/Conclusion

• Bender-Gestalt
• Constructs and models
• Previous research
• Overview of new developments
• Preliminary data analysis
• Conclusion
• Q A