Art Therapy

Journal of the American Art Therapy Association

ISSN: 0742-1656 (Print) 2159-9394 (Online) Journal homepage: http://www.tandfonline.com/loi/uart20

The Effectiveness of Art Therapy Interventions in

Reducing Post Traumatic Stress Disorder (PTSD)
Symptoms in Pediatric Trauma Patients

Linda Chapman MA, ATR-BC , Diane Morabito RN, MPH , Chris Ladakakos
PhD , Herbert Schreier MD & M. Margaret Knudson MD

To cite this article: Linda Chapman MA, ATR-BC , Diane Morabito RN, MPH , Chris Ladakakos
PhD , Herbert Schreier MD & M. Margaret Knudson MD (2001) The Effectiveness of Art Therapy
Interventions in Reducing Post Traumatic Stress Disorder (PTSD) Symptoms in Pediatric
Trauma Patients, Art Therapy, 18:2, 100-104, DOI: 10.1080/07421656.2001.10129750

To link to this article: http://dx.doi.org/10.1080/07421656.2001.10129750

Published online: 22 Apr 2011.

Submit your article to this journal

Article views: 6222

View related articles

Citing articles: 44 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=uart20

Download by: [UCLA Library] Date: 27 January 2017, At: 18:43

The Effectiveness of Art Therapy Interventions in
Reducing Post Traumatic Stress Disorder (PTSD)
Symptoms in Pediatric Trauma Patients
Linda Chapman, MA, ATR-BC, San Francisco, CA, Diane Morabito, RN, MPH,
San Francisco, CA, Chris Ladakakos, PhD, Oakland, CA, Herbert Schreier, MD,
Oakland, CA, and M. Margaret Knudson, MD, San Francisco, CA
Abstract
Although post traumatic stress disorder (PTSD) in children has
been extensively studied during the past 15 years, little research exists
regarding the efficacy of treatment interventions. This report describes
an outcome-based art therapy research project currently conducted at
a large urban hospital trauma center. Included are the theoretical
rationale and overview of an art therapy treatment intervention
called the Chapman Art Therapy Treatment Intervention (CATTI)
designed to reduce PTSD symptoms in pediatric trauma patients.
Used in this study, the CATTI was evaluated for efficacy in measur-
ing the reduction of PTSD symptoms at intervals of 1 week, 1 month,
and 6 months after discharge from the hospital. An early analysis of
the data does not indicate statistically significant differences in the
reduction of PTSD symptoms between the experimental and control
groups. However, there is evidence that the children receiving the art
therapy intervention did show a reduction in acute stress symptoms.
Introduction
In the United States, pediatric trauma is the leading cause of
death after 1 year of age. Each year, over 2 million children
require hospitalization after sustaining traumatic injuries. Over
100,000 children are permanently disabled, and 8,500 die each
year from accidental injuries. It is estimated that 300 children are
wounded each day by firearms, and five are murdered. In addi-
tion to the psychological costs to the family and society incurred
by this epidemic, over $15 billion is spent annually to care for
injured children (Baker & Waller, 1989; Deal, Gomby, Zippiroli,
& Behrman, 2000).
Physical traumatic injuries generate emotional phenomena
that include a defensive behavioral pattern, a grief reaction, and
a psychological retreat. Additionally, some patients exhibit PTSD
symptoms including reexperiencing the event, avoidance of
thoughts and feelings about the event, and hyper-arousal in the
Editors note: Linda Chapman, MA, ATR-BC, is an Investigator,
San Francisco Injury Center, University of California San Francisco.
Diane Morabito, RN, MPH, is Research Coordinator, San Francisco
Injury Center, University of California San Francisco. Chris Ladakakos,
PhD, is a Research Psychologist, Department of Psychiatry, Childrens
Hospital Oakland in Oakland, CA. Herbert Schreier, MD, is Chief,
Department of Psychiatry, Childrens Hospital Oakland, Oakland, CA.
M. Margaret Knudson, MD, is Director of the San Francisco Injury
Center, University of California San Francisco. All correspondence
should be sent to the first author at: 10151 East Road, Redwood Valley,
CA 95470 or redwood@itsa.ucsf.edu.
100
presence of relevant stimuli (American Psychiatric Association,
Diagnostic and Statistical Manual of Mental Disorders IV, 1994).
To address these symptoms in a population of hospitalized
children and adolescents in an urban Level I Trauma Center, a
program directed by the first author was established in 1988 by
the Departments of Pediatrics and Psychiatry at the University of
California San Francisco at San Francisco General Hospital. The
program offered psychological treatment services as a standard
component of care for the 1,200 patients age 0-18 admitted
annually. Seventy percent of the patients were trauma patients,
victims of gunshot or stab wounds and other acts of violence,
severe child abuse, or unintentional injuries such as motor vehi-
cle accidents, falls, and burns. The other 30% were admitted for
acute and chronic childhood illnesses such as asthma, sickle cell
disease, and neglect.
Children were seen beginning in the intensive care or pedi-
atric ward, and were offered individual trauma resolution thera-
py, group therapy when applicable, milieu activity therapy, and
Child Life-oriented services throughout their hospital stay.
With very limited opportunity for follow-up psychotherapy
services for the patients, it was hoped that early trauma resolution
therapy during hospitalization would reduce PTSD symptoms
and behaviors that may lead to disruptions in development,
future psychopathology, or recurring injuriesthe proposed but
not systematically tested results of untreated PTSD (Hubbard,
Realmuto, Northwood, & Masten, 1995; Tellez, Mackersie,
Morabito, Shagoury, & Heye, 1995; Terr, 1990).
History of the Current Project
Art therapists observations of hospitalized, traumatized chil-
dren indicated behaviors consistent with the DSM-IV definition
of PTSD and the criteria used for this diagnosis. Re-experiencing
phenomena were observed in post traumatic play, psychobiologi-
cal reactions, and recurrent intrusive images. Avoidance symptoms
were exhibited in gaze aversion, withdrawal, dissociative episodes,
apathy toward primary caregivers, refusal to comply with treat-
ment plans, and regressed developmental skills. Symptoms of
increased arousal were observed in nightmares and sleep distur-
bances, irritability, new aggressive behaviors, exaggerated startle
responses, and anxious attachment. The DSM-IV suggests that
these symptoms must be present for a period of 1 month or longer
to qualify for the diagnosis.
As part of the programming at San Francisco General
Hospital, an art therapy treatment intervention was developed to
reduce PTSD symptoms in patients prior to their discharge from
 CHAPMAN / MORABITO / LADAKAKOS / SCHREIER / KNUDSON
the hospital. Though not in a controlled study or systematic
appraisal, the intervention was implemented over a 10-year peri-
od with hundreds of hospitalized pediatric trauma patients across
the developmental spectrum with results that warranted a con-
trolled study. After nearly 10 years of programming, a research
proposal was submitted to the San Francisco Injury Center
(SFIC) for Research and Prevention, one of 10 Injury Control
Centers funded by the National Centers for Disease Control and
Prevention (CDC) in Washington, DC.
In a collaborative effort with Childrens Hospital Oakland in
Oakland, California, the project was expanded to include several
specific investigations. The specific aims of the study were:
To determine the incidence rate of PTSD in pediatric trau-
ma patients following an acute traumatic injury
To determine the outcome of a specific art therapy treat-
ment intervention in reducing PTSD symptoms in a popu-
lation of hospitalized children
To determine the outcome of the intervention at 1-week,
1-month and 6-month follow-up evaluations
To examine the ability of parents or caregivers to assess the
childs level of stress
To examine the ability of nursing staff to assess the childs
level of stress
This report features the intervention component of the proj-
ect, as other aspects of the study are beyond the limitations of a
brief report. Other publications covering other specific aims of
the project are currently in preparation.
Method
The study is a prospective, randomized cohort design. Sub-
jects were children 7 to 17 years old who had been admitted to a
Level I Trauma Center for traumatic injuries. The sustained
injury must have been of sufficient severity to require the child to
remain at the hospital for a minimum of 24 hours. Children were
excluded if admitted for injuries resulting from burns, child
abuse, and severe head injury. Severe head injury was defined as
an Abbreviated Injury Scale 3.
Unlike the adult literature, child PTSD literature indicates
that the type and severity of the trauma may affect the incidence
of PTSD. Therefore, an attempt was made to create a sample of
similarly traumatized children. Non-English-speaking children
and parents/caregivers were not included in the study due to a
lack of non-English-speaking therapists and interviewers.
Assessment and Follow-up
Tools and Procedures
The following assessment measures were used to assess base-
line measures of PTSD symptoms in all children, parents, and
caregivers enrolled in the study.
The measurement used to pretest and posttest the study
participants was The Childrens Post Traumatic Stress Disorder
Index (PTSD-I), designed and validated by Robert Pynoos, MD
(1997) for the assessment of post traumatic stress reactions of
school age children and adolescents after exposure to a broad
range of traumatic events. The PTSD-I is a self-report measure
that asks the individuals to respond to a 20-item inventory of
symptoms based primarily on the diagnostic criteria in the
DSM-IV. The PTSD-I also includes an adolescent and parent
101
version to assess parents reports of their childrens trauma symp-
toms. The PTSD-I scores are categorized according to the fol-
lowing breakpoints: raw score 12, doubtful PTSD; score of 12-
24, mild level of PTSD; score of 25-39, moderate level of PTSD;
score of 40-59, severe PTSD; score 60, very severe PTSD.
Interviews are repeated 1 week after injury, 1 month after hospi-
tal discharge, and 6 months after hospital discharge. One year
interviews are currently being conducted for children who con-
tinue to demonstrate symptomatology at 6 months. The age
range for the PTSD-I spans from age 5 to adulthood.
Baseline Measures
University of California at Los Angeles Post Traumatic Stress
Disorder Index (PTSD-I) Child or Adolescent Version
(Rodriguez, Steinberg, & Pynoos, 1997)
University of California at Los Angeles Post Traumatic Stress
Disorder Index (PTSD-I) Parent Version (Rodriguez,
Steinberg, & Pynoos, 1997)
Post Traumatic Stress Disorder Diagnostic Scale (Foa, 1995)
Family Environment Scale (Moos & Moos, 1994)
Nursing Checklist (three shifts) (modified version of PTSD-
RI-Parent Version)
Postintervention Measures
The same measures, with the exception of the Nursing
Checklist, were administered to all children and parents or
caregivers enrolled in the study at 1-week, 1-month, and 6-
month intervals.
Procedurally, all children admitted to the trauma service were
screened within 24 hours of admission. The childrens families
were offered a description of the project and an explanation of the
assessment, treatment, and follow-up procedures. The interviewer
obtained permission and proper documentation from child and
parent or caregiver to participate in the study, and the baseline
measures were completed by the parent and child. Those children
scoring 12 on the PTSD-I were randomized to a control group
or the intervention group. The control group received usual hos-
pital care. The intervention group received the art therapy inter-
vention from one of two participating art therapists. A group of
children with no PTSD symptoms were also assessed and followed
to determine a possible delay in symptom presentation.
Chapman Art Therapy Treatment
Intervention (CATTI)
A specific drawing treatment intervention was developed by
the first author as a brief, trauma resolution method. The Chap-
man Art Therapy Treatment Intervention (CATTI) is designed for
incident-specific, medical trauma to provide an opportunity for the
child to sequentially relate and cognitively comprehend the trau-
matic event, transport to the hospital, emergency care, hospitaliza-
tion and treatment regimen, and posthospital care and adjustment.
As the literature suggests, in cases of severe PTSD, incident-
specific, individual treatment is indicated (Frederick, 1985;
Shelby, 2000), and trauma mastery must be incident-specific for
specific resolution of the problem (Frederick, 1985, p. 93). The
CATTI offers developmentally and culturally sensitive trauma
resolution therapy to reduce PTSD symptoms, and to facilitate
the integration of the experience into ones larger, autobiograph-
102 ART THERAPY AND PTSD
ical life narrative (Siegel, 2000). Although children may benefit
from short-term therapy for incident-based trauma, many chil-
dren can be assisted with very brief, direct intervention (Shelby,
2000). In particular, art therapy facilitates the expression and
exploration of traumatic imagery as it emerges from memory and
finds form (Appleton, 2000).
The CATTI was developed by combining what is known
about the resulting neurochemical dysregulation response to
severe stress (van der Kolk, 1996), the formation of images and
their relationship to thought and cognition (Bruner, 1964;
Horowitz, 1970; Levick, 1984; Lusebrink, 1990), and develop-
mental, psychological, and neuropsychological theories on the
storage and retrieval of traumatic memories. The CATTI utilizes
the integrative capacity of the brain by accessing the traumatic
sensations and memories in a manner that is consistent with the
current understanding of the transmission of experience to lan-
guage (Horowitz, 1970; van der Kolk, 1994 ). By activating both
right and left hemisphere activity along with both visual and ver-
bal neural pathways, therapeutic potential is maximized as the
brain creates a visual, nonverbal narrative that is translated to a
coherent linguistic narrative. Coherent narratives require the par-
ticipation of both right and left hemispheres and are created
through hemispheric integration (Siegel, 1999).
The CATTI is facilitated in a one-to-one session at the
childs bedside in the hospital room. The CATTI is designed to
be completed in approximately 1 hour, with minimal art media.
Additional time may be required depending on the severity of the
event and meaning of the event to the child. Optimally, children
would receive ongoing psychological follow-up services, and min-
imally, at the time of medical follow-up care.
The protocol for this intervention is not described in this
paper, as some training is required to facilitate the CATTI. The
clinician must have knowledge of the psychological reactions to
traumatic physical injuries, the facilitation and containment of
traumatic imagery, and prior exposure to examples of a variety of
situations which may be presented (e.g., physical pain, psychobi-
ological reactions) and how to appropriately respond with art
media and imagery. The CATTI is currently being prepared for
publication, including the theoretical basis and practical applica-
tion for use with hospitalized pediatric trauma patients.
CATTI Procedure
The CATTI begins with a graphic kinesthetic activity
designed to stimulate the formation of images by activating the
cerebellum (Lusebrink, 1990). The motor activity allows for a
release of tension, and the rhythm and motion serve as a method
of relaxation (Lusebrink, 1990) to aid in tolerating the stressors
(Frederick, 1985). The kinesthetic activity is followed by a series
of carefully worded directives to elicit a series of drawings
designed to complete a coherent narrative about the event. Each
drawing in the narrative is followed by an opportunity for the
child to offer accompanying verbalizations related to each image.
The systematic method of accessing the material is consis-
tent with the childs visual and sensory modes of recording trau-
matic experiences (Terr, 1990; van der Kolk, 1994). The draw-
ings, or kinesthetic activity, activate the limbic system, the center
for emotional and perceptual processes. This is most often
expressed behaviorally in tears, or a loud, angry voice. The emo-
tional experience is considered to be essential in order for a cor-
rective emotional experience to occur (Schore, 1994). Childrens
emotional expressions are validated by the therapist as normal
responses to the traumatic event to universalize their experience
and reduce anxiety (Frederick, 1995).
The affective expression is often followed by a rapid shift to
cognitive functioning, evidenced in the childs ability to utilize
language and sequentially relate a coherent narrative about the
event and ensuing medical care. The child is also able to formu-
late adaptive coping mechanisms, and in some patients, the
orbital prefrontal system is engaged as the childs dialogue
includes more complex concepts such as consequences, compas-
sion, and a moral imperative.
Inherent in the CATTI is the opportunity for the child to
maintain control over the expression and containment of his or her
graphic and verbal traumatic material. The child is able to express
trauma-specific fears (Terr, 1990), perceptions, and misperceptions
about the event, medical care, and future. The child is also able to
relate his or her own unique story in a manner consistent with his
or her level of graphic, perceptual, and cognitive development.
After completing the drawing and verbal narrative, the child
is engaged in a retelling of the event, using the drawings to illus-
trate the narrative. During the retelling, numerous issues are
addressed, including but not limited to: misperceptions, rescue
and revenge fantasies, blame, shame and guilt, coping strategies,
treatment and follow-up plans, traumatic reminders, and reinte-
gration strategies.
Preliminary Results
A total of 85 patients were enrolled in the study: 31 received
the art therapy intervention, 27 received standard hospital care,
and 27 did not present PTSD symptoms at the baseline inter-
view. Hospital care was defined as the normal and usual course of
pediatric care including Child Life services, art therapy, social
work, and psychiatric consults.
Regarding child-specific variables, the average age of the
child was 10 years (mean 10.7, SD 2.6 years). The average
length of stay in the hospital was 4.4 days (mean 4.4, SD 5.5
days), and average length of stay in the ICU was 1.4 days (mean
1.4, SD 3.2 days).
Demographics
As illustrated in Table 1, ethnic representation is fairly well
mixed. The finding that 70% of the trauma victims were male is
consistent with adult trauma statistics. The most common mech-
Figure 1
Patient Groups
 CHAPMAN / MORABITO / LADAKAKOS / SCHREIER / KNUDSON 103

anism of injury is primarily from automobile vs. pedestrian or

automobile vs. bicycle accidents. Other injuries represented
include sports, animal, and penetrating injuries and falls. As
shown in Table 2, most of the injuries involved the extremities,
consistent with the high incidence of injuries resulting from
motor vehicle accidents in the sample.

CATTI vs. Standard Hospital Treatment

No significant overall difference in patients PTSD-I scores
was found between those who received the CATTI and those re-
ceiving standard hospital treatment. Table 1 illustrates the de-
crease in symptoms for each group of children from baseline to 1
week and 1 month. Although no statistically significant differ-
ence was determined, examination of individual symptom clus-
ters revealed that intervention produced a reduction in all DSM-
IV PTSD Criteria C (avoidance) symptoms at 1 week and sus-
Figure 2
tained that decrease 1 month. While this finding could be due to
Art Therapy vs. Standard Therapy
regression to the mean, it may indicate that the CATTI may be
effective in reducing acute stress symptoms, and may allow chil-
dren to discuss and process their traumatic experiences more interventions and strategies. This prospective research project
effectively when compared to the standard treatment group. offers one investigation that contributes to that effort.
There are several limitations to the study. First, most of the
Discussion children in the sample suffered mild to moderate injuries, and the
sample size is fairly small. Second, the assessment tools may not be
It is imperative for the art therapy profession to conduct sensitive enough to accurately measure symptoms of hospitalized
controlled studies to measure the efficacy of art therapy treatment acutely injured children. The presentation of post traumatic stress

Table 1
Demographics

Characteristic Percent
Race/Ethnicity
White 35.3%
Black 29.4%
Asian 23.5%
Hispanic 11.8%
Gender
Male 70.6%
Female 29.4%
Mechanism of
MVA/MCA 23.5% Figure 3
Auto/Ped or Bike 47.1% Percent Change in PTSD Scores
Other 29.4%

Table 2
Demographics

Characteristic Percent
Parental Marital Status
Married 41.2%
Single 58.8%
Body Region of Most Severe Injury
Head 17.6%
Chest 5.9%
Abdomen 5.9%
Extremities 58.9%
Multiple 11.8% Figure 4
Percent Change in PTSD Scores
104 ART THERAPY AND PTSD

symptoms in acutely injured children is often idiosyncratic and
manifested in conjunction with physical injuries, painful medical
procedures, and hospitalization. The currently available reaction
indices do not entirely describe the observable, often brief and
severe stress symptom presentation in acutely injured children.
Third, the intervention was designed to be administered as
a component of ongoing psychological pediatric trauma care, and
previous interaction with the injured child and his or her family.
In the study sample, children were seen without prior interaction
with the therapist facilitating the intervention. Finally, it is diffi-
cult to ascertain whether PTSD symptoms measured in the fol-
low-up data are independent of other traumatic events that may
have occurred in the childs life since the intervention. Or, as the
literature suggests, children with multiple traumas may have less
PTSD, which may be a possible variable in the efficacy of the
intervention. Perhaps, as suggested in recent literature, anxiety
and depression may be more descriptive of a childs reaction to
trauma than PTSD (Pynoos, Steinberg, & Piacentini, 1999; Terr,
1999; Yule, et al., 2000).
Further outcome results of 6-month follow-up data are
being compiled. Other data will include the physical and neuro-
logical reactions during the intervention (eye contact, sighs, tears,
mention of pain, autonomic nervous system responses, and psy-
chobiological reactions), and the childs perception of the perpe-
trator (object, person, etc.). Data related to the drawings will
include the amount of time spent on each drawing in the narra-
tive, the main focus of the drawing and number of details drawn,
and the amount of assistance, if any, required from the therapist
to complete the narrative. Additionally, the drawings may con-
tain valuable developmental information about how children per-
ceive traumatic events, medical care, and rehabilitation. It is
hoped that the data from this component of the study will con-
tribute to the development of the required assessment and treat-
ment tools to provide the essential psychological component of
comprehensive pediatric trauma care.
References
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual of mental disorders (4th ed.). Washington, DC: Author.
Appleton, V. (2001). Avenues of hope: Art therapy and the resolution of
trauma. Art Therapy: Journal of the American Art Therapy Association
18(1), 6-13.
Baker, S., & Waller, A. (1989). Childhood injury: State-by-state mortality
facts. Baltimore: The Johns Hopkins Injury Prevention Center.
Bruner, J. (1964). The course of cognitive growth. American Psychologist,
19, 1-15.
Deal, L., Gomby, D., Zippiroli, L., & Behrman, R. (2000).
Unintentional injuries in childhood: Analysis and recommendations.
The Future of Children, 10(1), 4-22.
Foa, E. (1995). Post Traumatic Stress Diagnostic Scale Manual.
Minneapolis: National Computer Systems, Inc.
Frederick, R. (1985). Children traumatized by catastrophic situations. In
S. Eth and R. Pynoos (Eds.), Post-Traumatic Stress Disorder in
Children. Washington, DC: American Psychiatric Press, Inc., 73-99.
Hubbard, J., Realmuto, G., Northwood, A., & Masten, A. (1995).
Comorbidity of psychiatric diagnoses with post traumatic stress disor-
der in survivors of childhood trauma. Journal of the Academy of Child
& Adolescent Psychiatry, 34, 1167-1173.
Levick, M. (1984). Imagery as a style of thinking. Journal of the American
Art Therapy Association, 1(3), 119-124.
Lusebrink, V. (1990). Imagery and visual expression in therapy. New York:
Plenum Press.
Moos, R., & Moos, B. (1994). Family Environment Scale Manual. Palo
Alto, CA: Consulting Psychologist Press.
Pynoos, R., Steinberg, A., & Piacentini, J. (1999). A developmental psy-
chopathology model of childhood traumatic stress and intersection
with anxiety disorders. Biological Psychology, 46, 1542-1554.
Rodriguez, N., Steinberg, A., & Pynoos, R (1997). PTSD index for
DSM-IV instrument information: Child version, adolescent version,
parent version. Unpublished manuscript.
Schore, A. (1994). Affect regulation and the origin of the self: The neurobi-
ology of emotional development. Mahwah, NJ: Erlbaum.
Shelby, J. (2000). Brief therapy with traumatized children: A develop-
mental perspective. In H. Kaduson & E. Schaefer (Eds.), Short-term
play therapy. New York: The Guilford Press, 69-104.
Shelby, J., & Tredinnick, M. (1995). Crisis intervention with survivors
of natural disaster: Lessons from Hurricane Andrew. Journal of
Counseling, 73(5), 491-497.
Siegel, D. (In Press). The developing mind and resolution of trauma:
Some ideas about information processing and an interpersonal neuro-
biology of psychotherapy. In F. Shaprio (Ed.), EMDR and the paradigm
prism. Washington, DC: American Psychological Association Press.
Siegel, D. (1999). The developing mind: Toward a neurobiology of inter-
personal experience. New York: The Guilford Press.
Tellez, M., Mackersie, R., Morabito, D., Shagoury, C., & Heye, C.
(1995). Risks, costs, and the expected complications of re-injury.
American Journal of Surgery, 170(6), 660-663.
Terr, L. (1999). A PTSD spectrum: The example of Challenger. In
Scientific Proceedings of 46th Annual American Academy of Child &
Adolescent Psychiatry, Chicago, IL, 90.
Terr, L. (1990). Too scared to cry. New York: Harper & Row.
van der Kolk, B. A., McFarlane, A., & Weisath, L. (Eds.). (1996).
Traumatic stress: The effects of overwhelming experience on mind, body
and society. New York: Guilford Press.
van der Kolk, B. (1994). The body keeps score: Memory and the evolv-
ing psychobiology of posttraumatic stress. Harvard Review of
Psychology, 1(5), 253-265.
Yule, W., Bolton, D., Udwin, O., Boyle, S., ORyan, D., & Nurrish, J.
(2000). The long-term psychological effects of a disaster experienced
in adolescence: The incidence and course of PTSD. Journal of Child
Psychology and Psychiatry, 41(4), 503-511.

Horowitz, M. (1970). Image formation and cognition. New York:

Appleton Century-Crofts.