Tova Clinical Manual
Tova Clinical Manual
Test Of Variables of Attention (abbreviated T.O.V.A.) is a registered trademark of The TOVA Company.
No part of this manual may be distributed without permission of The TOVA Company.
i
Table of Contents Page
I Introduction 1
4 The T.O.V.A. 20
4.1 Construction of the T.O.V.A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2 Sensitivity and Specificity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.3 Test-Retest Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.4 T.O.V.A. Formulas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.5 Screening Version . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.6 Clinical Version . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.7 Administering the T.O.V.A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
ii
4.7.1 Pre-test Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
9 T.O.V.A. ® Protocols 48
III Appendices 70
iii
T.O.V.A. Clinical Manual 1 TERMS AND CONCEPTS USED IN THIS MANUAL
Part I
Introduction
The T.O.V.A.
The Test of Variables of Attention® is an objective, standardized, and highly accurate continuous perfor-
mance test (CPT) that is used to measure attention and impulsivity. The Visual T.O.V.A. and Auditory
T.O.V.A. are non-language-based, sufficiently long (21.6 minutes), computerized tests that require no left-
right discrimination or sequencing. Responses are recorded with a specially designed, highly accurate (±1
ms) electronic microswitch.
The T.O.V.A. measures key components of visual and auditory attention. These measurements are then
compared to those of a group of people with attention disorders and a group without. The T.O.V.A. can also
establish a personal baseline of attention for future comparison or track response to treatment, including
medication. It can even flag possible symptom exaggeration.
Attention, according to Merriam-Webster’s dictionary is a) the act or state of applying the mind to some-
thing, and b) a condition of readiness for such attention involving especially a selective narrowing or
focusing of consciousness or receptivity.
Attention is a mainstay of life and the ability to focus one’s attention has a determination on one’s
success in school, at work, and in relationships.
Off-task behavior is a descriptive term with no specific etiologic or diagnostic implications. It is used in
this manual to indicate that persons are not engaged in the appropriate or assigned task when it is
reasonable to expect that they should be. They are not on-task for reasons that may have nothing to
do with attention disorders.
Attention-Deficit Hyperactivity Disorder (ADHD) refers to a specific symptom complex defined in the
current manual of diagnoses, the Diagnostic and Statistical Manual of Mental Disorders, fourth edition
(DSM-IV).
Target symptoms and measures refer to the particular symptoms that are specified for treatment and the
particular measures being used to determine the effectiveness of that treatment. In this manual,
attentional characteristics (specifically those variables measured by the T.O.V.A.) and hyperactivity
are evaluated and treated separately.
2.1.1 Sub-Types
Predominantly Inattentive Type (314.00) – Must have six or more of the following symptoms for six months
or more:
• Often fails to give close attention to details or makes careless mistakes in schoolwork, work, and other
activities
• Often does not follow through on instructions and fails to finish school work, chores, or duties in the
workplace (but not due to oppositional behavior and not because of a failure to understand)
• Often avoids, dislikes, or is reluctant to engage in tasks that require sustained mental effort (such as
schoolwork or homework)
• Often loses things necessary for tasks or activities (e.g. toys. school assignments, pencils, books, or
tools)
Predominantly Hyperactive-Impulsive Type (314.01) – Must have six or more of the following symptoms
for 6 months or more:
• Hyperactivity
– Often fidgets with hands or feet or squirms in seat
– Often leaves seat in classroom or in other situations in which remaining seated is expected
– Often runs about or climbs excessively in situations in which remaining seated is expected (in
adolescents and adults, may be limited to feelings of restlessness)
– Often has difficulty playing or engaging in leisure activities quietly
• Impulsivity
– Often blurts out answers before questions have been completed
– Often has difficulty awaiting turn
– Often interrupts or intrudes on others (e.g., butts into conversations or games)
Combined Type (314.01) – Must meet criteria for both inattentive and hyperactive/impulsive types for six
months or more
ADHD Not Otherwise Specified (314.9) – Criteria for other subtypes are not met, but symptoms are judged
to interfere with the affected individual’s functioning for six months or more. This category typically used for
adults with ADHD.
2.1.2 Requirements
To qualify for the diagnosis of ADHD, the following criteria must be met:
• There must be clinically significant distress or impairment in social, academic, or occupational func-
tioning
2.1.3 Issues
• ADHD is really a symptom complex, not a disorder (that, by definition, must have a single, common
etiology and a predictable natural history and response to treatment).
• The assumption that hyperactivity and attention deficits are necessarily linked is misleading and an
artifact of equating symptom complexes with disorders. (This isn’t the only example in psychiatry and
psychology of a hypothetical construct being treated as though it had an independent existence in the
real world).
• The requirement that the onset be by seven years of age ignores some critical factors. As examples,
many non-hyperactive, inattentive children and children with strong external support systems are not
symptomatic until later.
• Since behavior is situationally specific, attention deficits may not be apparent in more than one setting.
Differences in setting (school, classroom, teacher, peers) may affect the presence of symptoms.
• Although the emphasis is on inattention, individuals with ADHD are highly variable in their attention
over time, and can hyperfocus.
• There is no mention of the status of the Central Nervous System; that is, brain damage can cause
ADHD in DSM-IV.
• Normal behavior
Age-appropriate behavior that is mislabeled, e.g., “active alert” children or unrealistic adult expecta-
tions of normal development
• Medications
Such as anticonvulsants, antihistamines, and psychodepressants that sedate or slow the brain
• Toxic conditions
Such as environmental exposures, drugs, or an illness
Brain Injury
• Stress
Resulting from emotional trauma and overwhelming situations
• Learning disabilities
One third of individuals with an attentional disorder also have a learning disability, and vice versa
• Psychiatric conditions
Such as PTSD, psychosis, bipolar or obsessive-compulsive traits/disorders, depression, PDD, ODD,
dementia, conduct disorder, and reactive attachment disorder and/or anxiety
Note: These causes are not mutually exclusive. As noted above, 30% of individuals with ADHD
(including adults) have a learning disability (and vice versa), and between 40-65% of substance
abusers have ADHD. In addition, untreated individuals with ADHD often develop low self-esteem,
depression, and acting out which may obscure the underlying ADHD.
• History: Nothing (not even the T.O.V.A.) replaces a detailed personal and family history.
interview of teacher or supervisor are very helpful, especially to prepare them for recommendations.
• Mental status examination/personality assessment: This helps identify comorbid and/or other
conditions (such as depression).
• Behavior ratings for children—including the Vanderbilt, ACTeRS, the ASEBA, and BASC 2—are
especially useful. Behavior ratings for adults include the BAARS-IV, the ASEBA, and the BAADS.
While behavior ratings are an important part of the diagnostic process, they are best used in conjunc-
tion with objective measures like the T.O.V.A. to minimize the effects of rater bias and overemphasis
of disruptive behaviors.
• DSM-IV Symptom checklists: These checklists help clinicians to thoroughly review all symptoms of
ADHD.
• T.O.V.A.: The T.O.V.A. objectively measures visual and auditory attention, variability, response time,
and impulsivity, which can be affected by many factors; however, the T.O.V.A. does not make a
diagnosis. The clinician needs to make use of the objective results in the context of the full clinical
picture. There will be more about this in the review of sensitivity and specificity.
Note: A comprehensive work-up that includes all or most of the components above may not be
feasible or cost-effective. The clinician must decide which steps are needed and in what sequence.
Figure 1 illustrates the underlying neurophysiological problem in ADHD by comparing the subject’s original
baseline visual T.O.V.A. to his response to treatment—in this case, medication. The effects of 5 mg of
methylphenidate on the T.O.V.A. tests of a physician with an attention deficit are represented. Shown
are the mean response time histograms of the responses to the visual T.O.V.A. target stimuli. The mean
response time of 580 ms in the test without medication is significantly slower than the normal for an adult.
Indeed, 580 ms is normal for six-year-old boys. In contrast, the mean response time of 340 ms in the test
given 1.5 hours after 5 mg of methylphenidate is normal for an adult.
The treatment intervention is generally multifaceted, reflecting the many symptoms of ADHD and possible
comorbid conditions. Although the T.O.V.A. can be used to measure treatment effects on attention, variabil-
ity, response time, and impulsivity, we do not advocate treating the T.O.V.A. and losing sight of the person
in the real world.
A number of methods are currently being used to treat the symptoms of ADHD. These include:
• Providing information about ADHD and techniques to manage inattention and behavior
Confirming the diagnosis is sometimes sufficiently therapeutic, in and of itself. Even with the diag-
nosis, some cases are mild enough or the life circumstances (particularly for adults) are such that
education and awareness are the only treatment necessary. In some cases, other treatment may be
provided on an as-needed basis, for example when extensive reading is necessary.
• School/workplace consultation
Use of the T.O.V.A. Individualized Home and School Success Strategies can be very helpful in the
classroom setting and in the home. Findings need to be translated into relevant suggestions like a
private office space or a quiet homework area for a distractible person, etc.
• Vocational considerations
While we want to avoid self-fulfilling prophesies, attentional characteristics should be considered when
counseling about vocational opportunities. As an example, some persons with ADHD should be
advised to consider occupations that are active and varied, rather than tedious and repetitive. Some
individuals with ADHD tend to do well in a computer-related vocation.
• Recreational considerations
Consider faster-paced activities like swimming and handball instead of slowly paced activities.
• Neurofeedback
Using a pre- and a post-treatment T.O.V.A. (along with other measures), many clinicians working
with individuals with ADHD are reporting dramatic responses to these treatments. A number of well-
designed collaborative studies with the T.O.V.A. are now underway, and we expect to learn more about
the indications, types, and outcome of neurofeedback in the near future.
• Behavior modification
These types of treatment are extensively described in the literature. However, it should be noted that
they are especially useful in the treatment of impulsivity/disinhibition.
• Psychotherapy
This treatment is indicated with related self-defeating coping strategies and comorbid issues such as
depression, substance abuse, anxiety, ODD, and others.
• Natural Remedies
Continued research indicates there may be gains from Omega 3 fatty acids, vitamins, homeopathic
remedies, natural supplements, meditation, martial arts, brain development programs and exercises,
and others. Use of the T.O.V.A. provides an objective measure to the effectiveness of the treatment.
• Medication
See the section below.
Note: It is beyond the scope of this brief introduction to give extensive details about the pharma-
cotherapy of children and adults. For more information we recommend having and reading one of
the handbooks written on this subject.
As with any intervention, it is important to determine exactly what symptoms to target for treatment, and to
select appropriate measures of these symptoms. With the availability of CPTs like the T.O.V.A., the clinician
can now specifically target attentional variables and reliably, objectively measure medication effects. Be-
havior ratings alone may not be sufficiently sensitive enough to be used to determine optimal dosage levels
in the treatment of inattention.
Although antidepressants may be prescribed to treat ADHD, especially in cases with comorbid substance
abuse, they may not be as effective as stimulant medication on attention. The T.O.V.A. is used to measure
individual performance on antidepressants as well as with stimulants.
• Methylphenidate is the most frequently prescribed stimulant medication. Since the strength of the
generic (MPH) form can be ±20% of Ritalin™ , switching from one to the other can be problematic.
Ritalin SR (sustained release)™ and Ritalin LA (long acting) often has an uneven release over time
in our experience. In contrast, Concerta™ has a much smoother release over time and is particu-
larly useful when taking a noon dose is inconvenient (e.g., having to go to the school nurse’s office)
or problematic (e.g., forgetting to take the medication). Focalin™ (with liquid and tablet forms) and
Daytrana™ (the patch) appear quite promising, especially when small doses are prescribed.
• The recently approved Vyvanse™ is a “pro-drug” that is converted into an active stimulant in the
• Magnesium pemoline (Cylert™ ) is no longer generally available because of serious side effects (such
as liver damage).
NOTE: Since most recommendations about dosing are based on behavior ratings and body weight,
dosing may be excessive, particularly if the targeted symptom is inattention.
Bothersome but not serious side effects can include decreased appetite, weight loss, abdominal pain, sleep
disturbance (or increase), and afternoon irritability (“rebound”), which are often handled by a reduction in
dosage, administration of medication just before meals, or the addition of a smaller dose in the afternoon to
reduce “rebound”. Irritability, tiredness, or feeling “jumpy” or “edgy” are frequently encountered side effects
when a teenager or adult is given too much medication.
Less common effects may include tics, hypertension and tachycardia, temporary height suppression, and
sadness or depression, for which reassessment of the medication and dosage are promptly indicated. The
use of stimulant medication in persons who also have a tic or seizure disorder is controversial but is effective
in many cases.
Comparing the baseline (no medication) T.O.V.A. and a T.O.V.A. 1.5 to 2.5 hours after a single dose of
medication, one can predict whether there will be a good response to medication and determine a good
beginning dosage.
However, since patterns of behavior tend to change slowly, a clinical trial of medication is needed to ascer-
tain the degree of response. It is not possible to predict response to antidepressants in the same way since
they can take 3-4 weeks to build up.
A double-blind clinical study of the effects of MPH on attention and impulsivity was conducted with 154
children with ADHD. 143 children were MPH responders and 11 were non-responders.
While both groups showed some improvement in T.O.V.A. performance with MPH, only the responders
obtained a significant improvement (Figure 2).
With the availability of the T.O.V.A., the effects of different dosages can be very accurately measured to
obtain the best effects on attention. Use of the T.O.V.A. in clinical practice has resulted in a dramatic
increase in efficacy (95%), decrease in dosage levels (50%), low incidence of side effects (<5%), and
higher compliance. The decrease in dosage reflects the literature indicating that low doses of stimulant
medication affect attention, that high doses affect behavior, and that there is either little overlap between
the two effects, or that attention actually worsens with high doses (Figure 3).
In treating attention disorders, it is important to carefully determine target symptoms and the means of mea-
surement. For example, low-dose methylphenidate (MPH) affects attention primarily, while higher doses
affect behavior primarily and may affect attention adversely. If one were only to use subjective sources of
data (i.e.: history and behavior rating scales) to determine response to treatment, the effects on attention
could be missed.
• Atomoxetine (Strattera™ ) appears primarily useful in the treatment of problems of executive function.
Use the T.O.V.A. to compare on- and off-medication responses.
• Guanfacine (Intuniv™ ), an antihypertensive medication, was recently approved for use in ADHD, and
initial studies are encouraging.
• Wellbutrin™ and other antidepressants are prescribed for ADHD in cases with depression and/or
concern for potential substance abuse with stimulant medication.
• When a stimulant medication is used to treat an attention problem, Clonidine™ or Depakote™ may
be considered in addition.
Since optimal medication dosage can change over time, children and teenagers may need to be re-
evaluated every six months and adults once a year. It is recommended that the clinician obtain periodic
interim history (including undesirable treatment effects), parent/classroom/self behavior ratings, a new base-
line (no medication) T.O.V.A., and a new on-medication T.O.V.A. Evaluate results and determine whether to
continue or change treatment.
Below are two illustrations of the use of the T.O.V.A. and behavior ratings to monitor clinical course over
time. In the first case of a child diagnosed with ADHD, she gradually improved, and her T.O.V.A. and
behavior ratings normalized by age 10 1/2 (Figure 4). The two subsequent evaluations, which documented
her improvement, were obtained because of a research protocol. As would be true for any person who
“outgrows” ADHD, the last two off-medication T.O.V.A.’s were within normal limits, and the on-medication
T.O.V.A.’s improved but not significantly.
Figure 4: Outgrowing the need for medication: on- and off-medication trials over time.
The second illustration documents the clinical course of another girl diagnosed with ADHD who did not
improve by 11 1/2 (Figure 5). She may prove to be one of the 50% who do not “outgrow” her ADHD.
Figure 5: Continued need for medication: on- and off-medication trials over time.
3.1 History
Rosvold and his group introduced the CPT in the mid-1950s. His CPT was a sequential, visual, language
based A-X task in which the subjects responded whenever they saw an “A” followed by a “X”. Since that
time, many CPTs have been created primarily for use in research projects, but a few have been made
available commercially for researchers, schools, and clinicians. Research with early CPTs did not show
very promising results. These CPTs focused on omission and commission scores with inaccurate measures
of response time. We know that accurate Response Time (RT) and Variability of RT are critical for CPTs
to be sensitive and useful. Only the T.O.V.A. tests have been extensively normed and have highly accurate
(pm1 ms) and sensitive response time measures.
The T.O.V.A. began as a large electronic rack with a tachistoscopic shutter in our first clinical study in 1966.
It was nicknamed “Herman” by one of the children. With an accuracy of pm100 ms, the CPT results were
significant and documented the efficacy of stimulant medication (dextroamphetamine) in comparison with a
tranquilizer (chlorpromazine) in the treatment of hyperkinetic children. It is noteworthy that the classroom
behavior rating (the Conners Parent-Teacher Questionnaire) was not useful in discriminating medication
effects. Perhaps the most important outcome of this initial study was the necessity to target inattention and
hyperactivity separately, and to develop appropriate tools to measure each.
With the advent of the Apple IIe in the late 70s, the current design (with two test conditions, see below) and
the research-quality microswitch were created. It was initially named the “MCA” (or Minnesota Computer
Assessment). However, a potential copyright conflict arose, and the MCA became the Test Of Variables of
Attention (T.O.V.A.). As the T.O.V.A., it was normed and used in a number of clinical trials before its release
in the mid-80s.
Since then, we have continued to upgrade the test by making it more user friendly, collecting additional
subjects to have year by year norms for children, and adding signal detection indices for a comparison
ADHD score. The School Intervention Report and Version 7.0 with an improved scoring and interpretation
system were completed in 1996 when the Auditory T.O.V.A. was released. Additional norming studies and
programmatic enhancements, including the Home Intervention Report, led to the release of Version 7.3 in
2007. In 2008, the Symptom Exaggeration Index was developed. In 2011, version 8 was released with
several changes and new hardware. The Symptom Exaggeration Index was embedded in the report, the
ADHD score was reworked and became the Attention Performance Index (API) and additions were made
to the Home and School Intervention Reports which were renamed the Individualized Success Strategies-
Home and School.
Fourth-generation CPTs, like the T.O.V.A., accurately measure far more significant variables of both audi-
tory and visual information processing than the earlier CPTs. Length of the test (or subtest) makes a big
difference since some individuals with ADHD can “rise to the occasion” and do well enough for a short time.
Different CPTs may label these variables somewhat differently, making comparisons difficult. It is also true
that even when variables have the same labels, the characteristics of the different CPTs may be so different
that they are actually measuring very different things. Of course, the variables also have different values
within a CPT when there are two or more test conditions such as infrequent and frequent target presen-
tation modes, or when the inter-stimulus interval changes. In addition, we must keep in mind that labeling
something doesn’t mean that the variable is actually measuring what we think that it’s measuring.
The following variables are important for CPTs to measure, and T.O.V.A. includes them:
Response Time Variability is a measure of variability (consistency) of response time and is the standard
deviation of correct response times (that is, the time in ms taken to respond correctly to a target). Indi-
viduals with ADHD tend to be inconsistent—they may be able to perform within normal limits for a while,
but they “lose it” much sooner than others. As parents frequently note, a child with ADHD can focus (even
“hyperfocus”) and stay on task some times, particularly when the task is very interesting and fast paced (like
a computer game). Since Response Time Variability is the single most important measure of the T.O.V.A.
(accounting for >80% of the variance), the timing measurements must be very accurate; hence, the need
for an accurate microswitch rather than rely on the significantly less accurate mouse button or keyboard.
Correct Response Time is the processing time (in ms) taken to respond correctly to a target. Counter-
intuitively, individuals with ADHD often have slower than normal response times as well as faster ones. This
measure is one of the more important ones in the T.O.V.A., especially in the first (or boring) half, accounting
for >12% of the variance.
d0 or Response Sensitivity (the ratio of hit rate to false alarm rate) is a measure derived from Signal Detection
Theory. It is a measure of performance decrement, the rate of deterioration of performance over time. Most
individuals tend to fatigue over time, especially with a boring task. The performance of individuals with
ADHD deteriorates faster than others. d0 helps to distinguish individuals with ADHD from those without
ADHD and accounts for 6% of the variance in the T.O.V.A.
Errors of Commission are a measure of impulsivity and/or disinhibition and occur when the subject incor-
rectly responds to the nontarget; that is, the subject pushes the button when they shouldn’t have. In the
T.O.V.A., commission errors are far more frequent in the second half (high response demand). Since exces-
sive errors of commission can affect the other variables, they are also an important measure of test validity.
Generally, excessive commission errors decrease omission errors, shorten response times, and increase
variability. When a report states that the results are ‘invalid’ because of excessive commission scores, it
means that we must interpret the results cautiously since the other variables may or may not be valid. Of
course, excessive impulsivity is a hallmark of ADHD.
Errors of Omission are a measure of inattention and occur when the subject does not respond to the
designated target; that is, the subject omits pressing the button when a target appears or is sounded.
Because the T.O.V.A. covers a broad age span (4-80+), omissions in the visual (but not auditory) test have
a ceiling effect in adults. That is, the task is too easy for adults without ADHD who rarely make omission
errors. On the other hand, omission errors are a sensitive measure in children, teenagers, and the elderly.
When evaluating omissions, always look at the absolute or raw numbers. In some cases one or two errors
reach statistical significance yet there may be little or no clinical significance. As an example, a single error
early in quarter 1 may signify that the subject was surprised when the test began even though the practice
session preceded the test.
It is important to review the Tabulated Data form of the T.O.V.A. report to identify how many omission
errors occurred. Always interpret standard score (or standard deviation) data along side actual raw data to
determine clinical significance of the results.
Anticipatory Responses (AR) are a measure of impulsivity, guessing which stimulus is presented, or of a
different game strategy in which the subject may be trying to “kill” any stimulus as soon as possible. An
AR occurs whenever a response (pressing the microswitch) is made between 150 ms before and 150 ms
after any stimulus (target or non-target) appears or, in the case of the Auditory T.O.V.A., any stimulus is
heard. Most persons need more than 150 ms to distinguish between a target and a non-target and to
respond by pressing the microswitch; hence, the use of the word “anticipatory”. ARs are not included in the
calculations of errors, response times, and variability. Since excessive anticipatory responses can affect
the other variables, they are also an important measure of test validity. Generally, excessive anticipatory
responses decrease omission errors, increase commission errors, shorten response times, and increase
variability.
To prevent what may be incorrect interpretations, the T.O.V.A. Summary page labels the variables in any
quarter with excessive ARs (equal or exceeding 10%) as possibly “invalid” and needing to be cautiously
interpreted.
While we don’t want to confuse you (any further), it now turns out that there is a fourth reason why some
people have excessive ARs. Some people are much, much faster than most people. They are so fast that
they can accurately respond to the targets in less than the usual 150 ms, avoiding the non-targets. Thus,
when you examine the ratio of target to non-target ARs, you’ll find that these people have very few non-
target ARs. Most if not all of their ARs are with targets. Since the presentation of stimuli is randomized,
they can’t be guessing. They are really processing the information and responding significantly faster than
others.
As you might expect, some experienced computer game players, musicians, and athletes can perform so
well that their correct responses can fall into the AR range, and their test results are labeled as possibly
invalid by the interpretation program because of the excessive ARs.
Recognizing that some tests with excessive ARs should not be invalidated, we recommend that when there
are excessive ARs that the clinician examine the target : non-target ratio for ARs (Tabulated Data), and not
invalidate those quarters in which the ratio is better than 1 target : 3.5 non-targets in quarters 1 and 2 or
better than 3.5 targets : 1 non-target in quarters 3 and 4. In other words, a really fast and accurate person
will perform better than the ratio of targets to non-targets in that quarter.
Post-Commission Response Time is the response time immediately following a commission error. Clinical
observations (but not carefully conducted research) indicate that most people (including individuals with
ADHD) recognize when they make a commission error, and slow down for the next response. It is note-
worthy that a group of conduct disordered youngsters (without ADHD) either did not slow down or actually
responded faster than their average response time. Thus, this may be a way to distinguish individuals with
ADHD only from individuals with a conduct disorder only, but not the comorbid condition. Rarely, some
highly motivated individuals increase their focus, speed up after a commission error, and become more
accurate. A post T.O.V.A. interview with the subject may help to clarify the reason for fast post-commission
response times and adds depth to the clinical picture.
Multiple Responses are considered to be a reflection of neurological status. Excessive multiple responses
(>20/test) do not alter or invalidate the other variables, but they do appear to indicate nonspecific neurolog-
ical immaturity or dysfunction.
The following design features significantly influence what is being measured by a CPT as well as its “hit
rate”:
3.3.1 Stimuli
• Configuration
Simple stimuli (like in the T.O.V.A.) are easier to process than complex stimuli and have less associa-
tive value.
• Degradation
Although useful in work with schizophrenia, partial degradation of the stimuli are not features of CPTs
used to measure attention.
A fixed ISI (the interval between the stimuli) minimizes stimulating/alerting changes while a variable
ISI can be more arousing and/or difficult. The T.O.V.A. uses a two-second fixed ISI, which is generally
accepted as the most discriminating interval.
• Alerting signal
CPTs do not use alerting signals which would make the task easier and increase false negatives.
• Focal point
Focal points (like in the T.O.V.A.) are frequently used in visual CPTs.
The more complex CPTs can have significant practice effects, limiting their use as repeated measures. In
contrast, the T.O.V.A. actually has a small novelty effect—there may be non-significant commission error
changes (increases) in the first half of the second test but not thereafter. Thus, the T.O.V.A. can be repeated
even in the same day.
The longer the test, the harder it is to attend and inhibit. The 21.6 minute test with two 10.8 minute subtests
in the T.O.V.A. are in commercially available CPTs.
For children ages 4 and 5, the T.O.V.A. is 10.8 minutes in length, and the two subtests are each 5.4 minutes
each.
3.3.5 Norms
• Sample characteristics
Unlike the very carefully selected controls with no comorbidity for the T.O.V.A., some other CPTs have
inadequately defined and mixed samples.
• Controlled variables
Age: Since attention variables significantly change from birth to the late teens, year by year norms
are necessary for accurate measurement and comparison.
To illustrate, Figure 7 shows a norm curve of Total Response Time Variability (the first standard devi-
ation of response time) for females.
Figure 7: The response time variability for females in the T.O.V.A. norming study.
For the Visual T.O.V.A. there are year by year norms for each gender from 4-19 and grouped norms
by gender from 20-80+. For the Auditory T.O.V.A. there are year by year norms for each gender from
4-19 and grouped norms by gender from 20-29. The norms for the Visual T.O.V.A. and the Auditory
T.O.V.A. can be found in the T.O.V.A. Professional Manual.
Gender: Since males and females generally have significantly different norms, it is necessary to
have samples of each in the norms.
Intelligence: Intelligence may affect CPT performance and may be an important consideration. Re-
search on this topic is inconclusive.
Test conditions: Time of day and sequencing are important variables that can significantly affect
performance on a CPT. All norms for the T.O.V.A. were obtained in the morning, and the T.O.V.A. was
administered before other testing to avoid excessive fatigue.
3.3.6 Distractions
Few commercially available CPTs have distractors at this time. It is very difficult to control the novelty
(arousing) effects of so-called distractions which may actually enhance or decrease performance in some
cases. Some people come with their own built-in distractors (foot tapping, talking, chewing gum, etc) that
may act to arouse the person and help them focus. This warrants being noted.
The T.O.V.A. Microswitch is used to ensure that the timed responses are measured accurately. Small
variations and inconsistent timing can produce high false positives. To demonstrate why the T.O.V.A. uses
the microswitch, the following comparisons were made:
Thus, the microswitch in combination with T.O.V.A. software is significantly more accurate than the Conners’
or any of the other CPTs that use the mouse or keyboard and that use more than one computer. (ADHD
Report. 1995:3(6), 7-8).
4 The T.O.V.A.
The T.O.V.A. consists of two subtests with no transition or warning between them.
In the first half of the test ( the “Infrequent” or vigilance mode), the target appears randomly and infre-
quently with a target : non-target ratio of 1:3.5 The person presses the microswitch infrequently during this
quickly boring half. (There are 36 targets and 126 nontargets per quarter in quarters 1 and 2. Easily bored
(“low arousal”) persons may do poorly during this half.
In the second half of the test (the “Frequent” or high response demand mode) the target appears randomly
and frequently with a target : non-target ratio of 3.5:1. The person is frequently pressing the microswitch
and must inhibit the tendency to respond. (There are 126 targets and 36 nontargets per quarter in quarters
3 and 4.) Easily overstimulated (“high arousal”) persons may do poorly.
For children 4 and 5 years of age – the ratios of targets to non-targets remain the same; however, the
number of stimuli and length are half of those above.
The sensitivity of a test is its ability to correctly identify true cases (or, for the T.O.V.A. to identify ADHD);
the higher the sensitivity, the fewer false negatives (missing cases of ADHD).
The specificity of a test is its ability to correctly identify individuals who do not have ADHD. The greater the
specificity, the fewer false positives (incorrectly concluding that a person has ADHD).
Since sensitivity and specificity may vary inversely, it is necessary to arbitrarily select a cut-off that best
meets the expected use of the test. For the T.O.V.A., sensitivity and specificity were balanced to obtain
the highest accuracy of both rather than favor one or the other. For instance, we could have selected a
higher sensitivity for the screening version to minimize false negatives; however, the specificity would be
correspondingly lower, and there would be more false positives. Instead of using different cut-offs for the
two versions, they both have the same cut-off.
Of the two common ways to determine sensitivity and specificity, discriminant analysis is the usual and
less conservative one. If dealing with large representative samples from which generalizations about other
samples could be safely made, discriminant analysis would be the acceptable procedure. However, none
of the currently available CPTs, including the T.O.V.A., have large enough samples to use this statistical
method. We feel that it is best to be cautious about generalizing from one sample to others, and we do not
use discriminant analysis in determining sensitivity and specificity. If we did, the results would be:
• Discriminant analysis of T.O.V.A. variables with 29 UADD (ADHD without hyperactivity) cases and
29 matched controls correctly classified 79% of the UADD cases and 90% of subjects without ADHD.
Discriminant analysis of T.O.V.A. variables with 73 ADHD subjects and 73 matched subjects without
ADHD correctly classified 84% and 89%, respectively (see Table 2). ADHD and UADD subjects per-
formed more slowly and inconsistently and had more errors of omission (inattention) and commission
(impulsivity) than subjects without ADHD. Discriminant analysis of the T.O.V.A. and 10-item Conners’
Parent-Teacher Questionnaire of 61 of the youngsters with ADHD and 61 of the matched subjects
without ADHD correctly classified 87% of subjects without ADHD and 90% of the ADHD subjects with
13% false positives and 10% false negatives. Our sensitivities have been independently validated.
(See Forbes, G. B., Clinical Utility of the Test Of Variables of Attention (TOVA) in the Diagnosis of
Attention-Deficit /Hyperactivity Disorder. Journal of Clinical Psychology, Vol. 54 (4), 1998, 461-476.)
• Receiver Operator Characteristic (ROC) analysis is another way to determine sensitivity and speci-
ficity. ROC analysis is the more conservative technique that is used when it is best to be cautious
about generalizing from one sample to others. Even though to our knowledge the T.O.V.A. is the best
normed CPT, we should be hesitant to assume that the norms from a middle socioeconomic, pre-
dominant culture sample would apply to other samples everywhere even though the growing number
of culturally diverse norms that are appearing in the literature are remarkably consistent. While the
T.O.V.A. norm groups are noteworthy because of the absence of confounding comorbidities, they are
restricted- but perhaps less so than the usual samples of multi-problematic individuals. Accordingly,
the T.O.V.A. uses the more conservative ROC analysis (see Table 3).
Thus, there is an 80% chance that any given T.O.V.A. (with no other information about the individual) will
correctly identify whether the subject has ADHD or not. Of course, with the necessary clinically relevant
data (including behavior ratings and history), the “hit” rate improves significantly to above 90%.
Temporal stability of the Visual T.O.V.A. was examined in a study of school aged subjects using an interval
of ninety (90) minutes. The subjects were without histories of learning disabilities, psychiatric (including
ADHD) disorders, neurological disorders, or medical disease, and not on any medications.
The 24 subjects (15 males, 9 females) had a mean age of 8.31 years (SD = 2.35). Each subject received
an initial T.O.V.A. and was re-administered the second T.O.V.A. ninety minutes after completing the initial
test. Each subject completed both tests by 1 p.m. on the same day.
Data analysis yielded no significant differences within the group between the two tests (Table 2).
Table 4: Table of Means, SD, t values and Correlations for Test / Retest Scores
The Visual T.O.V.A. demonstrated temporal stability over a ninety minute episode. The ninety minute test
interval was selected because many clinicians use the T.O.V.A. as a tool to determine the effects of stimulant
medications by comparing a baseline (no medication) test and an on-medication test 90 minutes later.
Another test-retest study with 33 randomly selected non-ADHD children, 40 children with ADHD, and 24
non-ADHD adults also revealed no significant differences in the T.O.V.A. variables, using paired t-tests.
Formulas and algorithms for calculating the all T.O.V.A. variables can be found in the T.O.V.A. Professional
Manual.
The Screening Version of the T.O.V.A. is used by schools, other educational settings (including learning
centers), and non-licensed health care professionals. Uses of the Screening Version include:
The Screening Version is the same test as the Clinical Version; however, the printouts are formatted dif-
ferently to reflect that they are used for screening rather than as part of a clinical assessment or treatment
monitoring process.
The Screening Version Summary states that it is used for screening purposes only, not for clinical assess-
ments. The results are or are not within normal limits, and if not within normal limits, a referral to a clinician
is warranted. Contact the T.O.V.A. Referral Service and telephone number, 1-800-REF-TOVA (733-8682)
for referrals to clinicians specializing in attentional disorders in that area.
The Screening Version includes the Individualized Success Strategies-School and Home
The Clinical Version of the T.O.V.A. is used by licensed clinicians and researchers.
The Clinical Form analyzes the Raw Data and includes the Individualized Success Strategies-School and
Home.
• Measuring attention in individuals being evaluated for an attention disorder, including ADHD and
Traumatic Brain Injury
• Measuring attention in other disorders affecting the brain, including AIDS and dementia
Training of both professionals and nonprofessionals to administer and monitor the test should follow the
general outline of the instructions in the Professional Manual and include the use of the T.O.V.A. ® Rating
Form for recording observations that may be helpful to the clinician. In general, we want the subject to
balance speed and errors—to be as fast as they can be, yet to minimize errors. (See Appendix A for Testing
instructions.)
The T.O.V.A. should be administered in the morning to comply with the norming procedure and to min-
imize diurnal variability which can significantly affect test performance. (When comparing two tests it is
especially important that they have been given at the similar time of day (that is, both in the morning or, if
necessary, the afternoon.) If the T.O.V.A. is part of a battery of tests, it is important to administer it first—
before the subject is fatigued or bored. If both the auditory and visual T.O.V.A. are to be administered, a
sufficient time (>1.5 hours) should elapse between the tests to enable the subject to rest.
The norms were obtained with an observer present at all times in the room with the subject. Research
has shown that the observer’s presence makes a significant difference even though they are not interacting
with the subject. Test performances by children and adults can be significantly worse when the observer is
not present.
When testing for the first time, the practice test should be given in its entirety. For subsequent testing, only
a portion of the practice test may be necessary to remember the task.
Although prompting is helpful in the practice test, it is not used during the actual testing unless absolutely
necessary since prompting was not given for the norms.
The T.O.V.A. ® Rating Form (Appendix A) can be used to record observations during testing. This form is
not copyrighted so that it can be duplicated and used as needed.
Preparing the subject is crucial, because it assures that the test is administered properly and replicates the
same conditions as the T.O.V.A. norms.
1. Prior to testing, explain to the subject (or to caregivers) that no caffeinated beverages (e.g. coffee, tea,
cocoa, or soft drinks) should be ingested on the day of a test. Nor should the subject have smoked.
2. Time of Day: Testing is best done between 6 AM and 1 PM to control any diurnal effects.
3. Setting: Testing should be done in a quiet, darkened room with a glare-free monitor screen. Clocks
should not be visible or audible. It is best if the subject faces a neutral colored wall without distracting
pictures. The monitor should be placed so that the keyboard is not visible or available to the subject.
4. At test time:
• Introduce yourself to the subject.
• Ask if the subject needs to use the bathroom.
• Determine whether they have glasses or hearing aid if needed.
• Have subject remove his or her watch.
• Determine from subject or caregiver any medications taken in the last 12 hours (Ritalin, anticon-
vulsants, and/or antihistamines). Add these, with dosage and interval since administered, in the
New Test Session window.
• Position the subject and chair so he or she may sit with feet on the floor.
• Position the monitor so the screen is at or near eye level.
Part II
The clinician determines the application of the T.O.V.A. in measurement, tracking, assessment and treat-
ment. Familiarity with the scientific basis of the T.O.V.A. as well as the interpretation of the T.O.V.A. report
will help establish the best use of the T.O.V.A. Thousands of clinicians have used the T.O.V.A. for a variety
of purposes that call for the objective measurement of variables of attention. The following information is
based on both the readout of the reports and the clinical work of the authors. Clinical casework or reflec-
tions are chosen to illustrate the authors’ use of the T.O.V.A. and are not a recommendation for assessment
or treatment.
The T.O.V.A. performance can be significantly improved or worsened by anything that affects attention:
• Someone with ADHD could successfully self-medicate with nicotine and/or caffeinated beverages,
assuming that excessive quantities are not ingested. A person with ADHD who has coffee, an energy
drink, or cigarettes before testing may very well perform within normal limits on the T.O.V.A.
• On the other hand, acute caffeine and nicotine withdrawal can have adverse effect on attention. Thus,
a person without ADHD can do poorly on the T.O.V.A. if they do not indulge in their habitual two or
more cups of coffee in the morning.
• Any medication that can affect brain function can affect attention. Someone taking antihistamines for
allergies can become sufficiently sedated so that the T.O.V.A. performance may not be within normal
limits just as someone receiving lithium for a bipolar disorder may have slow response times.
• People with ADHD who have extensive video game experience and highly trained athletes, etc. may
perform normally on the visual T.O.V.A. due to the eye-hand training. The auditory T.O.V.A. is useful
in these situations although musicians may do better on the auditory T.O.V.A. than others.
• Sleep deprivation, anxiety, and depression, as well as a number of psychiatric conditions, can ad-
versely affect performance whether comorbid with ADHD or not.
• Although the literature is not definitive, a person with an above average intelligence may perform
better than average. Conversely, someone with a lower than average intelligence may perform less
than average.
It is important that the clinician obtain an adequate history to be able to interpret T.O.V.A. results,
taking these factors into account.
The following pages contain a Clinical T.O.V.A. Interpretation Report and discussion of the forms and find-
ings. Some of the forms are designed for use by researchers and some by advanced T.O.V.A. users. When
printing out a Report, you may select which pages you want printed and not print out the other pages. When
sending a Report to a non-clinical setting such as a school, we recommend sending only the Introduction
and the Summary forms. When sending a Report to a clinician, we recommend sending the Introduction,
Summary, Analyzed Data, Analyzed Data Page, Tabulated Data, and Raw Data Graphs.
Introduction
This page provides basic information about the T.O.V.A. and its uses.
Summary
This page and pages to follow include Demographic data about the subject—ID number, subject’s name,
gender, and age—as well as general session information—visual or auditory, version #, serial #, date and
time of administration.
• T.O.V.A. Interpretation: summarizes the results by comparing the subject’s performance with indi-
viduals who do not have an attention problem and with individuals who have been independently
diagnosed with ADHD, matched for age and gender.
The interpretation statement will be one of the following:
– If the performance and the Attention Performance Index (API) are within normal limits:
The results of this T.O.V.A. test are within normal limits.
– If the performance is not within normal limits but the API is within normal limits, the sentence will
be:
The results of this T.O.V.A. test are not within normal limits and are suggestive of an
attention problem.
– If the performance and the API are not within normal limits, the sentence will be:
The results of this T.O.V.A. test are not within normal limits and are suggestive of an
attention problem, including ADHD.
– If the performance is within normal limits but the API is not within normal limits, the sentence will
be:
Although the results of this T.O.V.A. test are within normal limits, based on the pattern
of performance there is some evidence of a possible attention problem, including ADHD
(see the Attention Performance Index score).
Note: “Suggestive of an attention problem” does not mean that the person has ADHD, but
only that the results were not within normal limits.
Since the performance and the API were not within normal limits in the sample protocol, the
Interpretation is:
The results of this T.O.V.A. test are not within normal limits and are suggestive of an
attention problem, including ADHD.
• Session and Response Validity summarizes whether there were any occurrences that might in-
validate or effect the results, such as user interruptions, excessive errors, or the testing was not
administered in the morning.
When one or more quarters are labeled “invalid”, and the remaining quarters are within normal limits,
the protocol would be interpreted as within normal limits. When this happens, a statement is made
that the invalid quarter(s) may have been the result of an attention disorder, and that the printed
interpretation and results should be viewed cautiously.
Validity Measures
– Tests obtained after 1 PM must be interpreted cautiously since all of the norms were obtained
T.O.V.A. ® (Visual)
Age Quarters 1 or 2 Quarters 3 or 4
T.O.V.A. ® (Auditory)
• Symptom Exaggeration Index (SEI) is a measure of symptom exaggeration or “faking bad”. The
SEI is only applicable for ages 17 and older and relevant when the overall performance is not within
• Treatment: Current treatments, including any prescribed or over the counter medications (with dosages
and medication-test interval), if any, are recorded.
Note: The clinician will need to determine what effects the treatment may have on the T.O.V.A.
performance. For the sample protocol:
None was entered.
– The first statement summarizes whether the overall performance (the T.O.V.A. Interpretation)
was or was not within normal limits. For the sample protocol:
The overall performance was not within normal limits.
– The next statements summarize which variables, if any, were not within normal limits.
For the sample protocol:
Inconsistency (Response Time Variability) was borderline in Q1 & Q2 and not within
normal limits in Q3 & Q4. This finding is important since inconsistency is the most
sensitive measure in the T.O.V.A. Impulsivity (Commission Errors) was not within normal
limits in Q3.
Comment: Note that when Response Time slows the most in quarter 4, Commission Errors
significantly improve.
– Bar graphs
The quarter by quarter standard scores for the four primary variables are illustrated.
Standard scores above 110 are above average.
Standard scores between 85-110 are average.
Standard scores 80-85 are borderline.
Standard scores below 80 are not within normal limits.
If the standard score is below the limit of the vertical axis (<40), it would be noted as a downward
facing red triangle.
If the standard score is above the limit of the vertical axis (>120), it would be noted as an upward
facing red triangle.
Analyzed Data
* [ ] (score in brackets) means that the quarter may not be valid and must be interpreted
cautiously.
* ! ! (score between two exclamation points) means that there were excessive Commission
Errors in that quarter. Quarters with excessive Commission Errors need to be interpreted
cautiously since Response Time Variability can be artificially increased, and Response Time
and Omission Errors can be.
* * (score flagged with an asterisk) means that the results are valid, not within normal limits,
and compatible with an attention disorder.
• The Graphs present T.O.V.A. results using standard scores. If the standard score is below the limit of
the vertical axis (<40), it would be noted as a downward facing red triangle.
The “X” axis of the graph will be moved (up or down) automatically as necessary to allow sufficient
room for the results. If the standard score is above the limit of the vertical axis (>120), it would be
noted as an upward facing red triangle.
The results of four tests can be displayed side-by-side to compare performances over time or no
treatment and treatment. See the User’s Manual for details on how to compare tests.
110 110
100 100
Standard Score
90 90
80 80
70 70
60 60
50 50
40 40
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
83 83 71 63 116 106 105 92
110 110
100 100
Standard Score
90 90
80 80
70 70
60 60
50 50
40 40
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
85 89 78 104 100 100 100 100
T.O.V.A. Clinical Manual 6 THE T.O.V.A. REPORT
Tabulated Data
As noted Tabulated Data is provided for researchers and advanced T.O.V.A. users.
• Comments, if any.
Data are listed by quarters, halves and total, and include:
– RT Variability (ms)
– Response Time (ms)
– Post-Commission Responses (#, Response Time in ms, and Variability in ms)
– D Prime1 (Standard Score and Raw Score)
– Commission Errors (# and %, ), Response Time (ms)
– Omission Errors (# and %)
– Anticipatory Responses (To Nontargets- # and To Targets- #)
– Multiple Responses #
– Total Correct # (Correct Responses # Correct Nonresponses #)
– Beta1
– User Interrupts #
– Hardware Errors #
• Session Parameters
– Standard configurations:
Infrequent = quarters with infrequent targets
Frequent = quarters with frequent targets
IIFF for ages 6- to 80+ (Standard or form 1 with 4 quarters, 21.6 minutes)
or
IF for ages 4 and 5 (form 2 with 2 quarters, 10.8 minutes)
– Inter-Stimulus Interval (time between successive stimuli): 2000 ms
– On-Time (time when the stimulus appears in a 2000 ms period): 200 ms
– Off-Time (time when the stimulus disappears in a 2000 ms period): 300 ms
– Anticipatory Cutoff: If a response (microswitch press) occurs within 150 ms of On-Time (200 ms),
the response is considered an Anticipatory (usually a guess or impulsive response)
Note: Research configurations may be altered as needed
No tester identification*, designation of test, or comments were recorded. Test Format was
#1 (IIFF with the standard timing). The test (version 7.2) was administered on May 31, 2005
and reformatted in version 8.0 on July 23, 2010.
* The tester identification was removed for the sample protocol.
Codes:
• Light gray area = Responses within normal limits for age and gender.
750
600
Q1 450
300
150
0
-150
0 40 80 120 160
Trial #
1,050
900
Milliseconds
750
600
Q2 450
300
150
0
-150
0 40 80 120 160
Trial #
1,050
900
Milliseconds
750
600
Q3 450
300
150
0
-150
0 40 80 120 160
Trial #
1,050
900
Milliseconds
750
600
Q4 450
300
150
0
-150
0 40 80 120 160
Trial #
T.O.V.A. Clinical Manual 6 THE T.O.V.A. REPORT
In addition to the identifying information, these tables present the sequence of targets and non-targets, and
the subject’s response to each one. Errors are shown in red, and response times are in ms. A negative
response time indicates a response that was made before the stimulus appeared.
Error Graphs
This error descent graph displays the number of Commission, Omission, and Anticipatory Errors and norms
by quarter and total errors (the red line). The gray area represents the norm (standard score of 100) for age
and gender. The light gray line represents a standard score of 85, and the black line represents a standard
score of 80. We are in the process of determining the clinical usefulness of this display.
625 625
Correct Responses
Correct Responses
600 600
575 575
550 550
525 525
500 500
625 625
Correct Responses
Correct Responses
600 600
575 575
550 550
525 525
500 500
T.O.V.A. Clinical Manual 6 THE T.O.V.A. REPORT
This histogram is included for researchers and advanced T.O.V.A. users. We are examining new mod-
els of Reaction Time that may make the T.O.V.A. even more accurate and helpful for clinicians and re-
searchers. We have been using the traditional Gaussian distributions (“bell curves”) to model Reaction
Time histograms. We are now exploring a new model, the “ExGaussian” distribution with its unique mea-
sure, Tau.
Gaussian and ex-Gaussian models are very similar, except for Tau. The ex-Gaussian Mu corresponds to
the Gaussian Mu (the average response time), and ExGaussian Sigma corresponds to the Gaussian Sigma
(variability or width of the curve). The new parameter, Tau, models an exponential decay of the Reaction
Time histogram to the “right” (longer reaction times). Normally Tau is very small; however, recent research
indicates that Tau may be a more sensitive parameter defining attention deficits than either Mu or Sigma
alone. Our hypothesis is that although Mu and Sigma may be within normal limits, a significantly skewed
Tau may be indicative of ADHD.
Codes:
The ms for Mu, Sigma, and Tau are listed by quarters, halves, and total in the table.
80
75
70
65
60
Number of responses
55
50
45
40
35
30
25
20
15
10
0
0 250 500 750 1,000 1,250 1,500 1,750
Reaction time (ms)
These strategies are designed for individuals who need support in their school environment. Whether the
person has an attention problem, demonstrates behavior problems, or just needs more structure to succeed,
these strategies focus on creating successful habits and developing successful structure in the classroom
and school environment. Interpreting a student’s T.O.V.A. results along with interviews and behavior ratings
can help determine the most appropriate strategies for the individual. We recommend coordinating the
school and home strategies for the student to maintain consistency and minimize the possibility of overload.
Available by clicking on the Help button on the main T.O.V.A. Screen, these strategies are also an entry into
dialogue with teachers and school staff.
These strategies are designed for individuals who need support in their home environment. These may be
used for children, adolescents and adults. Depending upon the person’s age, they may also be adapted
to be utilized in the workplace if needed. If the individual is in school, we recommend coordinating the
home and school strategies to ensure consistency in the approach to success. Whether the person has
an attention problem, demonstrates behavior problems, or just needs more structure to succeed, these
strategies focus on creating successful habits and developing successful structure in the home environment.
Interpreting a person’s T.O.V.A. results along with interviews and behavior ratings can help determine the
most appropriate strategies for the individual.
9 T.O.V.A. ® Protocols
Reading and understanding the T.O.V.A. in the context of the a clinical evaluation or in establishing a base-
line for attention requires an understanding of the relationship between the variables and the quarters, as
well as the meaning of the API and the SEI and their link to the Comparative to the Normative Sample
standard scores. When first reading a T.O.V.A. follow the standard scores for each variable from quarter
to quarter. Look for scores below 85 to indicate difficulty with consistency, processing speed, impulsivity
or attention. Note decreases or increases between quarters and halves. Pay attention to the relationship
between certain variables. For instance, is the person fast but makes a lot of errors, or are they slow and
make few errors? Is the person inconsistent in the first half (low arousal) but fast, consistent and accurate
in the second half (high arousal). This can tell you something about the person and the person’s test taking
strategy. One can extrapolate from this to how the person may behave, but it is important to ask the subject
if your “story” fits. Often it will if you follow the inferences that one can deduce from the T.O.V.A.
Experienced T.O.V.A. users consult the Tabulated Data, the Error Graph and the Raw Data Graphs pages
on a regular basis to gain more insight into the subject’s performance on the T.O.V.A., and the Raw Data
Tables when necessary. Each of these pages give more data that help “tell the story” of the subject.
Along with the four variables, the T.O.V.A. also provides information on vigilance over the 21.6 minutes,
adaptability (changes from half 1 to half 2), and neurological status (consecutive omissions and excessive
multiple responses). Gathering all this data together helps form a picture of the person’s performance
on the T.O.V.A. which can be useful in understanding the person and in designing appropriate treatment
interventions or protocols.
The API is not a read alone score. It should always be analyzed in relationship to the Comparison to the
Normative Score. The API score is a comparison of the person’s scores based on selected measures that
persons with an independent diagnosis of ADHD frequently demonstrated. Primarily focused on variability,
response time and d’ prime (degradation of performance over time), the API score is an adjunct to the Com-
parison to the Normative Sample. It provides research-based information and can be used as additional
information in understanding the subject’s performance and when comparing tests pre- and post-treatment
to determine one aspect of the success of treatment.
The SEI is a research based measure determined by the results of persons who “faked bad” on the T.O.V.A.
If the SEI score indicates the possibility of malingering and secondary gain is a concern, consider the
possibility of “faking bad”, or other reasons that a person may have triggered a high score on the SEI,
such as drug use, psychiatric conditions, oppositional behavior and/or very severe problems with attention.
The SEI scores are consistent with people who fake bad on the test; however, some persons with severe
attention problems may trigger the SEI. The SEI is meant to alert the clinician to the possibility of “faking
bad” and to provide a window for consideration of secondary motivation as well as for conversation with the
subject concerning their test results.
Though two T.O.V.A.’s may look similar, remember that no two people are the same. Always understand
the T.O.V.A. in the context of the person, and the person in the context of the T.O.V.A. results. Good clinical
practice makes use of the T.O.V.A. in interpreting the person, but relies on a complete assessment of the
person to determine diagnosis and treatment decisions.
The protocols below are presented and reviewed to illustrate the T.O.V.A. s and their uses.
Clinical Protocol #1
This 41-year-old male has been in psychotherapy off and on for 7 years due to relational issues. Gradually,
the therapist learns that he frequently doesn’t keep commitments, avoids paper work, accounting, decision
making, and doesn’t pay attention to tasks that require sustained attention. He frequently forgets appoint-
ments and feels overwhelmed. No history of learning disabilities or psychiatric diagnoses. He recently
reported that he had a head injury in a motorcycle accident 20 years ago when he was unconscious for
unknown amount of time.
T.O.V.A. Interpretation: The results of this T.O.V.A. test are not within normal limits, and suggest the
presence of an attention disorder, including ADHD.
Comparison to the Normative Score: Good Response Time Variability (RTV) in low arousal condition
(Quarters 1 and 2). However, the standard score drops 8 points in the change to high arousal (Quarter 3)
and drops to 66 (not within normal limits) in Quarter 4. Response Time (RT) drops 8 points from Quarter
2 to Quarter 3 and then another 19 points to 89 in Quarter 4. Commission Errors are within normal limits.
Omission Errors drop to 56 (not within normal limits) in Quarter 3 and improve to 100 in Quarter 4. Overall,
the scores suggest difficulty with transition and adaptability (RTV, RT and Omissions drop in Quarter 3) and
tiring after 15 minutes (RTV and RT drop in Quarter 4).
Attention Performance Index score is 0.57 (within normal range). Comments: He was referred to his
personal physician with a preliminary diagnosis of Traumatic Brain Injury.
T.O.V.A. Interpretation: The results of this T.O.V.A. are within normal limits.
Comparison to the Normative Sample: All 4 Quarters are within normal limits.
Comments: T.O.V.A. and behavior ratings are within normal limits. Good response to medication.
Clinical Protocol #2
This 17-year-old female was referred because of academic difficulties. She reported being frequently bored,
daydreaming excessively and feeling mildly depressed.
T.O.V.A. Interpretation: The results of this test are not within normal limits, and suggest the presence of
an attention problem, including ADHD.
Comparison to the Normative Sample: The overall results are not within normal limits. Inconsistency
(Response Time Variability) was not within normal limits in Quarters 1 and 2 and was borderline in Quarter
3. This finding is important since RTV is the most sensitive measure of the T.O.V.A. Interestingly, her RTV
scores progressively improved throughout the test, so that by the end (Quarter 4) her RTV score was within
normal limits.
Attention Performance Index: 1.36 (within normal limits) Comments: Based on history, behavior ratings
and T.O.V.A. scores, she was referred to her physician who prescribed medication and requested an on-
medication T.O.V.A. The 18-year-old female (case # 2) was administered the T.O.V.A. on trial medication.
T.O.V.A. Interpretation: The results of the T.O.V.A. are borderline, and may suggest an attention problem,
including ADHD.
Comparison to the Normative Sample: The overall performance was borderline. Impulsivity was border-
line in Quarter 4. RTV and RT improved from baseline T.O.V.A.; however, Commission Errors were worse
as were Quarter 4 Omissions.
Attention Performance Index: 3.63 (unremarkable and better than baseline) Comments: The on-medication
results are mixed (some variables better and some worse) when compared with baseline T.O.V.A. API score
is much improved; however, this doesn’t take into account the borderline impulsivity score. Overall, mixed
results are usually indicative of excessive medication and that a lower dosage may produce better results.
See below.
Based on self-report and the mixed T.O.V.A. results with 5 mg of Adderall, the 18-year-old female (case #
2) was retested on a lower dosage of medication.
T.O.V.A. Interpretation: The results of this T.O.V.A. test are within normal limits.
Comparison to the Normative Sample: All quarters are within normal limits and as good or better than
baseline scores.
Attention Performance Index: 5.12 (better than baseline and with 5 mg of Adderall)
Comments: Good response to 2.5 mg of Adderall. API continues to improve and all quarters are within
normal limits.
Clinical Protocol #3
History: This boy, almost 11 years old, was referred for psychotherapy due to impulsive behaviors, dis-
tractibility, lack of focus, difficulty with follow through and arguing with his parents. He had a previous
diagnosis of ADHD, combined type, and was on medication at the beginning of treatment. At the time of the
referral the physician was considering raising his medication to obtain better clinical results.
T.O.V.A. Interpretation: The results of this T.O.V.A. are not within normal limits, and the overall pattern of
performance is suggestive of an attention problem, including ADHD.
Session and Response Validity: There are excessive >10% Anticipatory Errors in Quarters 3 and 4 (21%
and 26.5%, respectively). Since excessive ARs tend to increase Commissions and decrease Omissions,
the results need to be interpreted cautiously.
Symptom Exaggeration Index: Not available for a person under 17 years of age.
Comparative to the Normative Sample: The overall performance was not within normal limits. Response
Time Variability was borderline in Quarter 1 and not within normal limits in Quarters 2 and 4. This is
important as it is the most sensitive measure of the T.O.V.A. Response Time was not within normal limits in
Quarter 4. Impulsivity (Commission Errors) was not within normal limits in Quarters 1 and 2 and inattention
(Omission Errors) was not within normal limits in Quarter 4. There were 25 multiple responses in Quarter
4. Note that in Quarter 3, all variables were within normal limits. It may be that the increased arousal in
Half 2 improved performance, but after 5 minutes, he began to tire inQuarter 4 (increased Response Time
Variability, Omissions and Multiple Responses.).
API: -5.61 (compatible with a pattern commonly seen in persons diagnosed with ADHD)
Case #3: T.O.V.A. test to compare his baseline T.O.V.A. to his medication scores.
T.O.V.A. Interpretation: The results of this test are not within normal limits, and are suggestive of an
attention disorder, including ADHD.
Comparison to the Normative Sample: This test is not within normal limits. Impulsivity (Commission
Errors) was borderline in Quarter 2 and not within normal limits in Quarters 1 and 4.
Attention Performance Index: 1.27 (not compatible with a pattern commonly seen in persons diagnosed
with ADHD) Comments: The API score is much improved in comparison to the baseline. However, some
variables are now worse on medication than when off medication. Impulsivity is now worse and this is one
of the behaviors that the subject exhibits. Mixed results suggest a lower dosage of medication may work
better for this subject.
Case #3: Test to compare his baseline T.O.V.A. to his 50 mg Adderall XR T.O.V.A. score and his 10mg
Adderall XR T.O.V.A. score.
T.O.V.A. Interpretation: The results of this test are within normal limits.
Treatment: Challenge dose of 10mg of Adderall XR taken 2.5 hours before testing.
Attention Performance Index: 3.37 (better than baseline and with 50 mg of Adderall XR) Comments: All
scores are within normal limits. The API is higher than in both previous tests. Impulsivity scores are very
good. The dosage was decreased to 10 mg of Adderall XR rather than raising dosage from 50 mg.
Clinical Protocol #4
History: This 9-year-old girl was tested because of inattention and verbal impulsivity in the classroom where
she was not performing well. The same problems were evident in the home as well. A visual T.O.V.A. and
an Auditory T.O.V.A. were administered to determine her strengths and challenges with visual and auditory
processing.
T.O.V.A. Interpretation: The results of this T.O.V.A. are not within normal limits and the overall pattern of
performance suggests an attention problem, including ADHD.
Comparison to the Normative Sample: The overall performance was not within normal limits. Response
Time Variability (RTV) was borderline in quarters 1 and 2 and not within normal limits in quarter 3. Response
Time (RT) was borderline slow in quarter 1 and gradually increased in quarters 2, 3, and 4. Commissions (a
measure of impulsivity) were within normal limits in all four quarters but were significantly higher in quarter
3. Omissions (a measure of inattention) were within normal limits in Half 1 but were significantly deviant
(high) in Half 2. There is one set of three omissions in a row in quarter 4. As noted elsewhere in the
Manual, this could be indicative of an underlying neurological problem, such as narcolepsy or seizures, and
is reason for a referral to a neurologist. Overall, she had difficulty with consistency and speed in half 1 (low
arousal), had difficulty with impulsivity in the change to half 2(high arousal) and had difficulty with attention
throughout half 2 (high arousal).
Attention Performance Index: -2.59. Her performance was similar to that of someone with ADHD.
Note: Subject was retested one week later at a similar time of day on the Auditory T.O.V.A. to compare her
performance to her Visual T.O.V.A.
T.O.V.A. Interpretation: The results of this T.O.V.A. are not within normal limits and the overall pattern of
performance suggests an attention problem, including ADHD.
Comparison to the Normative Sample: The overall performance was not within normal limits. Her per-
formance on the Auditory T.O.V.A. was significantly worse than on the Visual T.O.V.A. for all four variables.
Interestingly, her Response Time was fastest in quarter 3 when she made the most Commission errors.
When she slowed in quarter 4, her Commissions improved. Omissions, which were worse in Half 1, oc-
curred in bunches (3-7 in a row) twice in quarter 3 and six times in quarter 4, rather than the usual scattered
pattern in an attention disorder. These are shown in the Raw Data Tables in the report. As noted else-
where in the Manual, this could be indicative of an underlying neurological problem, such as narcolepsy or
seizures, and is reason for a referral to a neurologist.
Notes: The Attention Performance Index research for the Auditory T.O.V.A. is pending.
Up to 10-12% of people process visual differently than auditory information; hence, as in this case, the
recommendation to do both Visual and Auditory Screening T.O.V.A.
The subject’s performance is not within normal limits on both the Auditory T.O.V.A. and the Visual T.O.V.A.
As the Auditory T.O.V.A. is worse, some clinicians use it as a baseline while others use the Visual T.O.V.A.
or both. Treatment, including school and home strategies (see Individualized Success Strategies-Home
and School), should take into account the need to address both sensory domains.
Part III
Appendices
For a copy of the T.O.V.A. observer Rating Form, please contact The TOVA Company.
2 Often fails to give close attention to details or makes careless mistakes in schoolwork, work, or other
activities.
2 Often loses things necessary for tasks or activities (e.g. toys, school assignments, pencils, books, or
tools).
Have these behaviors persisted for more than 6 months? YES NO Don’t know
Have these behaviors been maladaptive and inconsistent with development level? YES NO Don’t know
Were these behaviors present before age 7? YES NO Don’t know
Do these behaviors occur in more than one setting
(e.g. at home and at school or at home and at work)? YES NO Don’t know
Have these behaviors impaired social relationships? YES NO Don’t know
Have these behaviors impaired academic or work performance? YES NO Don’t know
Has there ever been any other psychiatric or psychological diagnosis before? YES NO Don’t know
If so, what and when?