Psychopharmacology (2011) 215:177–189
DOI 10.1007/s00213-010-2123-1
ORIGINAL INVESTIGATION
T-pattern analysis of diazepam-induced modifications
on the temporal organization of rat behavioral
response to anxiety in hole board
Maurizio Casarrubea & Filippina Sorbera &
Magnus S. Magnusson & Giuseppe Crescimanno
Received: 27 July 2010 / Accepted: 29 November 2010 / Published online: 23 December 2010
# Springer-Verlag 2010
Abstract
Rationale By means of t-pattern analysis, it has been
observed that the different events, characterizing rat
behavior in hole board (HB), present close interrelation-
ships which occur sequentially and with significant con-
straints on the interval lengths separating them.
Objectives The aim of present research was to study, by
means of descriptive and multivariate t-pattern analyses, the
effects of the reference anxiolytic drug diazepam (DZP) on
temporal structure of a rat’s anxiety-related behavior in HB.
Methods Fifty-six male Wistar rats were tested for 10 min
in HB. Video files, collected for each animal, were coded
by means of a software coder, and event log files, generated
for each subject, were analyzed by means of a specific
software for temporal pattern analysis (t-pattern analysis).
Results Significant diazepam-induced modifications were
observed for durations of walking, climbing, edge-sniff,
and face grooming. Dose-dependent decreases of t-patterns’
total amount, of their mean occurrences and of their mean
length for each group were detected. Also, t-patterns’ mean
occurrences, in terms of different composition, were
reduced. Percent distributions showed a significant increase
M. Casarrubea : F. Sorbera : G. Crescimanno (*)
Department of Experimental Biomedicine and Clinical Neurosciences,
Human Physiology Section, University of Palermo,
Palermo, Italy
e-mail: crescima@unipa.it
M. Casarrubea
e-mail: m.casarrubea@unipa.it
F. Sorbera
e-mail: sorbera@unipa.it
M. S. Magnusson
Human Behavior Laboratory, University of Iceland,
Reykjavik, Iceland
e-mail: msm@hi.is
of t-patterns including walking for all administered groups,
and significant reductions of t-patterns including climbing,
immobile sniffing, and edge-sniff. Front-paw licking and
face grooming were reduced at the higher DZP dose.
Conclusions Present study demonstrates, for the first time,
that the temporal structure of Wistar rats’ behavioral response
to anxiety in HB changes following pharmacological manip-
ulation of anxiety condition. Moreover, t-pattern analysis is
suggested to represent a useful tool to evaluate and compare
different classes of anti-anxiety molecules.
Keywords Anxiety . T-pattern analysis . Diazepam . Hole
board . Multivariate analysis . Rat
Introduction
Various features of anxiety-related behaviors in rodents
have been investigated by means of hole board (HB), an
experimental assay widely used in the last decades
(Adamec et al. 2006; Casarrubea et al. 2009b, 2010a; File
and Wardill 1975; Harada et al. 2006; Kamei et al. 2007;
Saitoh et al. 2006). Similar to open-field test, HB offers a
simple approach to the study of rodent behavior in aversive
situations such as an unknown environment (Choleris et al.
2001). Multivariate analysis, a set of techniques able to
provide information concerning multiple behavioral varia-
bles (Espejo 1997; Casarrubea et al. 2006, 2008, 2009c;
Cruz et al. 1994; Kalueff and Tuohimaa 2005; Lino-de-
Oliveira et al. 2005; Spruijt and Gispen 1984; van Den
Berg et al. 1999), has described rodent activity in HB in
terms of behavioral structure (Casarrubea et al. 2009a, b,
2010b) and its temporal organization (t-pattern analysis)
(Casarrubea et al. 2010a). In particular, by means of
t-pattern analysis, it has been observed that the different
178
events, characterizing rat behavior in HB, present close
interrelationships which occur sequentially and with signif-
icant constraints on the interval lengths separating them
(Casarrubea et al. 2010a). On this point, when behavioral
units are repeatedly performed, as parts of non-randomly
repeated real-time patterns, variations in the content and
structure of t-patterns, as a function of experimental factors,
can be expected to reflect changes in underlying neuro-
physiological processes. Accordingly, t-pattern analysis has
already been successfully utilized in previous researches to
study behavioral modifications in neuro-psychiatric con-
ditions (Kemp et al. 2008; Lyon and Kemp 2004), patterns
of behavioral synchronization in human beings (Grammer
et al. 1998), motor skill responses in body movement
(Castañer et al. 2009), hormonal–behavioral relationships
(Hirschenhauser et al. 2002), interaction among human
beings and animals (Kerepesi et al. 2005) and route-tracing
stereotypy in rodents (Bonasera et al. 2008).
The aim of the present research was to study, by means
of descriptive and t-pattern analyses, whether and how the
temporal structure of a rat’s behavior in HB is affected by
pharmacological manipulation of anxiety condition. To this
purpose, different doses of the reference anxiolytic drug
diazepam (DZP) were administered to different groups of
male Wistar rats.
Materials and methods
Apparatus
The HB apparatus consisted of a square arena 50×50 cm
made of white opaque Plexiglas with a raised floor,
positioned 5 cm above a white opaque Plexiglas sub-floor
and containing four equidistant holes, 4 cm in diameter.
Each hole center was 10 cm from the two nearest walls so
that holes were equidistant from adjacent corners. The
arena was surrounded by three white opaque Plexiglas
walls and a front transparent one (Casarrubea et al. 2009a).
Subjects and drugs
Fifty-six specific pathogen-free male Wistar rats (Morini,
Italy), 55–60 days old, weighing 220–250 g were used.
Animals were individually housed in a room maintained at
23±1°C, under a 12-h light/dark cycle (light on, 7:00 a.m.–
light off, 7:00 p.m.). Food (standard laboratory pellets—
Mucedola, Italy) and water were available “ad libitum”.
Animals were randomly assigned to four groups of 14
subjects: one group received 1 ml of saline IP as control
and three groups were IP injected with DZP (Roche; 0.25,
0.5 and 2 mg/kg in 1 ml of saline). DZP doses were chosen
since they have been demonstrated to provide a clear
Psychopharmacology (2011) 215:177–189
anxiolytic effect in rodents (lower dose=0.25 mg/kg)
without affecting locomotor activity (higher dose=2 mg/kg)
(Casarrubea et al. 2009b; Harada et al. 2006).
Experimental procedure
All experimental observations were carried out from 08:00
to 12:00 a.m. To minimize possible transfer effects and
avoid possible visual or olfactory influences, rats were
transferred from housing to testing room inside their own
home cages and allowed to acclimate for 30 min far from
the HB apparatus. Environmental temperature in the testing
room was maintained equal to the housing room. Each subject
was placed in HB and allowed to freely explore for 10 min.
Rats, experimentally naïve, were observed only once. After
each observation, HB apparatus was carefully cleaned with
ethylic alcohol (70%). Rodents’ behavior was recorded
through a digital camera, and video files were stored in a
personal computer for following analyses. The camera was
placed in front of the transparent Plexiglas wall.
Principles of laboratory animal care were followed, and
experiments were complied with the current laws of Italy
on the Protection of Animals. Experimental procedures
were approved by a veterinary committee appointed by the
University of Palermo and conducted in accordance with the
European Communities Council Directive (86/609/EEC).
Data analysis
The ethogram (Table 1), including 11 behavioral elements,
was the same utilized in our recent studies (Casarrubea et
al. 2009a, b, 2010a). Video files were coded by means of a
software coder (The Observer, Noldus Information Tech-
nology BV, The Netherlands) and event log files generated
for each subject. To detect temporal relationships among
behavioral elements, event log files were processed with
THEME™ (PatternVision Ltd, Iceland and Noldus Infor-
mation Technology BV, The Netherlands), a software
carrying out an analysis aiming at evaluating possible
significant relationships (i.e., critical intervals) among
events in the course of time (t-pattern analysis) (Magnusson
2000). T-pattern detection algorithm searches for relation-
ships between events in behavioral data by taking into
account, the order, timing, and frequency of these events.
Figure 1 exemplifies the occurrences of a three-element
t-pattern within a hypothetical data set encompassing 40
events. The algorithm compares the distributions of each
pair of the behavioral elements A and B searching for a
time window (interval) so that, more often than chance
expectation, A is followed by B within that time window. In
this case, A and B are by definition a t-pattern indicated as
(A B). In a second step, such first level t-patterns are
marked and considered as potential A or B terms in higher
Psychopharmacology (2011) 215:177–189 179

Table 1 Ethogram of rat behav-

ior in hole board Wa Rat walks around sniffing the environment
Cl Rat maintains an erect posture leaning against the Plexiglas box. Usually associated with Sniffing
IS Rat sniffs the environment standing on the ground
HD Rat puts its head into one of the four holes
ES Rat sniffs the hole border without inserting the head inside
Re Rat maintains an erect posture without leaning against the plexiglas box. Usually associated with sniffing
FPL Rat licks or grooms its forepaws
Wa walking, Cl climbing, IS HPL Rat licks or grooms its hind paws
immobile sniffing, HD head-dip,
ES edge-sniff, Re rearing, FPL FG Rat rubs its face (ears, mouth, vibrissae, and eyes) with rapid circular movements of its forepaws
front-paw licking, HPL hind-paw BG Rat licks its body combing the fur by fast movements of incisors
licking, FG face grooming, BG Im A fixed posture is maintained
body grooming, Im immobility

patterns, for example, ((A B) C). Thus, more complex
patterns may be created following this hierarchical bottom-
up detection process up to any level. When no more
patterns are found, the search stops. A more detailed
description of concepts and theories behind t-pattern
analysis, and related detection procedures, can be found in
our previous articles (Casarrubea et al. 2010a; Magnusson
2000, 2004).
T-pattern detection software THEME™ allows the
researcher to set specific search parameters. In the present
study, t-pattern analysis has been carried out by using the
same search parameters employed in our recent investiga-
tion (Casarrubea et al. 2010a):
– “significance level” (maximum accepted probability of any
critical interval relationship to occur by chance)=0.0001
– “lumping factor” (forward and backward transition
probability above which A and B of a t-pattern (A B)
are lumped, that is, A and B are not considered
separately but only as the (A B) pattern)=0.90
– “minimum samples” (minimum percent of subjects in
which a pattern must occur to be detected)=0.51
– “minimum occurrences” (minimum number of times a
t-pattern must occur to be detected)=8
The following parameters of behavioral response to
anxiety were studied: (1) mean durations of each behavioral
element and mean latencies of its first appearance; (2)
overall number of t-patterns detected for each group; (3)
mean number of detected t-patterns for each group; (4)
mean number of detected t-patterns, on the basis of their length
(that is, for each group, the mean number of detected two-,
three-, four-, and five-element t-patterns); (5) mean number of
different t-patterns detected in each group; (6) behavioral
density stripes; (7) percent modification of t-patterns including
each behavioral element; (8) percent modification of t-patterns
including specific behavioral sequences.
Graphical representations
Due to the large amount of space necessary to illustrate
individual t-patterns by means of classical tree representa-
tions (Fig. 1) (Magnusson 2000), t-patterns behavioral
stripes, that is, the graphical representation of t-patterns
occurrences in a compact way along a timeline x-axis, have
recently been introduced (Casarrubea et al. 2010a). Thus, in
the present paper, besides histograms showing mean values,
behavioral stripes are presented.
Statistics
Between groups comparison
One-way analysis of variance (ANOVA), followed by
Newman–Keuls post hoc test for multiple comparisons,
was carried out to assess possible diazepam-induced
modifications of: (a) each behavioral element’s mean
duration; (b) mean latency of each behavioral element’s
first appearance; (c) mean number of detected t-patterns for
each group; (d) mean number of detected t-patterns on the
basis of their length and (e) mean number of detected

Fig. 1 Exemplifying t-pattern encompassing three hypothetical events. behavioral occurrences (grey letters) are left out of the whole timeline.
Bold arrow, period of observation (T0–Tx) consisting of 40 events “First”, “second”, and “following steps” (ellipses) indicate search runs
(black letters). The ((A B) C) t-pattern becomes evident if all the other carried out by THEME software following a bottom-up detection process
180
t-patterns on the basis of their different composition. Chi-
square test was carried out to assess possible diazepam-
induced modifications on percent of t-patterns encompass-
ing each behavioral element and on percent of t-patterns
encompassing specific sequences.
Within-group statistical validation
Each t-pattern implies a statistical significance test among
critical intervals separating its structural elements. Howev-
er, the huge number of possible relationships raises the
question whether the number of detected t-patterns is
different by chance. THEME™ deals with this critical
issue by repeatedly randomizing and analyzing original
data. Thus, for each experimental group, the mean number
of t-patterns (±1 SD) detected in randomized data was then
compared with the one obtained from the real data.
Results
Durations and latencies
Mean duration±SE of each behavioral component is pre-
sented in Fig. 2. ANOVA showed significant diazepam-
Fig. 2 Mean durations±SE (in
seconds) of each behavioral
pattern. Data were obtained
from the analysis of four groups
of 14 rats IP injected with saline
or diazepam 0.25, 0.5, and
2 mg/kg dissolved in 1 ml of
saline. * p≤0.05, significant
differences in comparison with
saline (ANOVA and Newman–
Keuls post hoc test). For abbre-
viations, see Table 1
Psychopharmacology (2011) 215:177–189
induced modifications for walking (F3,963 =8.69; p≤0.0001),
climbing (F3,725 =10.99; p≤0.0001), edge-sniff (F3,496 =4.10;
p=0.007) and face grooming (F3,92 =3.07; p=0.032). New-
man–Keuls post hoc test for multiple comparisons highlight-
ed significant (p≤0.05) modifications for walking at all DZP
doses, climbing at DZP 0.25 and 2 mg/kg, edge-sniff at DZP
2 mg/kg and face grooming at DZP 0.5 and 2 mg/kg.
Latencies of each behavioral element’s first appearance
showed no significant diazepam-induced changes.
T-patterns number and length
DZP administration induced a dose-dependent decrease
of total detected t-patterns amount: 2,750 (saline), 2,289
(DZP 0.25 mg/kg), 1,937 (DZP 0.5 mg/kg) and 1,762
(DZP 2 mg/kg).
Mean number±SE of detected t-patterns in each group is
presented in Fig. 3. ANOVA showed a significant effect of
DZP (F3,55 =3.908; p=0.013). Newman–Keuls post hoc test
for multiple comparisons revealed significant differences
(p≤0.05) between saline and the two higher DZP doses (0.5
and 2 mg/kg).
Mean number±SE of detected t-patterns, on the basis of
their length, is illustrated in Fig. 4. ANOVA showed a
significant effect of DZP for all t-patterns of different
Psychopharmacology (2011) 215:177–189
Fig. 3 Mean number±SE of t-patterns for each group. *p≤0.05,
significant differences in comparison with saline (ANOVA and
Newman–Keuls post hoc test). Data were obtained from the analysis
of 56 rats
lengths except for the two-element ones (F3,55 =2.55; p=
0.06; F3,55 =17.38; p≤0.0001; F3,55 =3.14; p=0.033; and
F2,41 =7.81; p=0.001, respectively for two-, three-, four-,
and five-element t-patterns). Newman–Keuls post hoc test
revealed significant changes (p≤0.05) for three- and five-
element t-patterns at all doses and for four-element t-patterns
at 0.5 mg/kg dose.
Figure 5 illustrates the mean number±SE of different
t-patterns on the basis of their different composition. ANOVA
showed a highly significant effect of DZP (F3,55 =11.74;
p≤0.0001). Newman–Keuls post hoc test highlighted signif-
icant (p≤0.05) modifications for all doses of DZP.
T-patterns composition
Individual behavioral components
To illustrate the occurrences of the numerous and different
t-patterns during the 10-min period of observation through
their tree representation, a large amount of space would be
Fig. 4 Mean number±SE of
t-patterns on the basis of their
lengths. *p≤0.05, significant
differences in comparison with
saline (ANOVA and
Newman–Keuls post hoc test).
Data were obtained from the
analysis of 56 rats
181
required. A demonstration of this possible drawback is
represented in Fig. 6 showing trees of four different
t-patterns, each encompassing walking and concerning only
one subject. Taking into consideration that: (a) much more
different t-patterns, encompassing the same behavioral
component, are present in the same subject, (b) 11 different
behavioral components are present in the ethogram and (c)
14 subjects have been tested for each of the four different
groups, a comprehensive t-pattern description, represented by
behavioral stripes, has recently been introduced (Casarrubea
et al. 2010a). Figure 7 shows stripes of behavioral density on
the basis of the presence of t-patterns encompassing each
element of the ethogram (Table 1). Since lower level
t-patterns may be included in higher level ones, bar
thicknesses express behavioral density in terms of t-pattern
occurrences. The stripes show that the largest amount of
t-patterns includes walking, climbing, immobile sniffing,
head-dip, and edge-sniff. Such t-patterns were found
prevalently within the first 5-min observation. Rearing,
front-paw licking, face grooming, body grooming, and
immobility were less represented. No t-patterns including
rearing were detected at DZP 0.5 and 2 mg/kg whereas those
encompassing grooming were not identified with the higher
DZP dose. Table 2 presents quantitative aspects underlying
results illustrated in Fig. 7, that is, DZP effects on percent
distribution of t-patterns encompassing each behavioral
element. DZP induced a significant (p≤0.05, chi-square test)
increase of t-patterns including walking at all doses, rearing
at 0.25 mg/kg and immobility at 0.5 and 2 mg/kg.
Significant reductions were found at all doses for t-patterns
encompassing climbing, immobile sniffing, and edge-sniff;
as to front-paw licking and face grooming, a decrease was
observed at the higher DZP dose. Concerning t-patterns
including head-dip, significant reductions were detected at
DZP 0.25 and 0.5 mg/kg and significant increase at the
higher dose. Finally, t-patterns including body grooming at
182
Fig. 5 Mean number±SE of t-patterns on the basis of their different
composition. *p≤0.05, significant differences in comparison with
saline (ANOVA and Newman–Keuls post hoc test). Data were
obtained from the analysis of 56 rats
the higher dose and hind-paw licking at all doses were not
evidenced.
Behavioral sequences
Density stripes, on the basis of t-patterns containing
specific behavioral sequences are illustrated in Fig. 8. The
evaluated sequences were chosen taking into consideration
the already demonstrated relationships among their consti-
tutive components (Casarrubea et al. 2009a, b, c). For all
groups, t-patterns were prevalently found within the first
period of observation, except (face grooming ➔ body
grooming) sequence which presented a more homogeneous
distribution during the 10-min temporal window. In
addition, at the higher dose, no t-patterns encompassing
(face grooming ➔ body grooming) were detected. Table 3
presents quantitative aspects underlying results illustrated in
Fig. 8 and shows DZP effects on percent distribution of
t-patterns encompassing specific behavioral sequences.
At all doses, DZP induced significant (p≤0.05, chi-square
test) reductions of t-patterns containing (walking ➔ climb-
Fig. 6 Occurrences of four different t-patterns encompassing walking
along a 10-min observational period. Progressive numbers, inside
brackets on the left, indicate events order appearance within each
Psychopharmacology (2011) 215:177–189
ing) and (edge-sniff ➔ head-dip) sequences. Significant
increase of t-patterns including (climbing ➔ walking) and
(head-dip ➔ edge-sniff) were evidenced for all DZP
administered groups. T-patterns containing (edge-sniff ➔
walking) were significantly reduced at DZP 0.25 and
0.5 mg/kg and significantly increased at the higher dose.
T-patterns encompassing (head-dip ➔ walking) were signif-
icantly reduced at the lower DZP dose and significantly
increased at 0.5 and 2 m/kg. Finally, t-patterns containing
(face grooming ➔ body grooming) at the higher DZP dose,
and those encompassing (body grooming ➔ face grooming),
in control group and at all doses, were not detected.
Within groups statistical validation
Within groups comparison is illustrated in Fig. 9. For each
experimental group, t-patterns length distribution in real
data and in random generated data (±1 SD) is showed.
Results indicate that the largest amount of independent
t-patterns detected is present in real, rather than in random
generated data.
Discussion
The present study demonstrates, for the first time, that
pharmacological manipulation of anxiety condition causes
clear and significant modifications in the temporal structure
of a rat’s behavioral response to anxiety in hole board, both
in terms of complexity and variability.
Durations and latencies
Walking duration has been demonstrated to be significantly
increased following DZP (Fig. 2). Therefore, it would be
possible to conclude that anxiolytic treatment, making an
animal less anxious toward an unfamiliar environment,
facilitates a rodent’s exploratory propensity. However, if
t-pattern: (1), first event; (2), second event; (3), third event; (4), fourth
event; (5), fifth event. Data were obtained from the analysis of rat 1
from saline group
Psychopharmacology (2011) 215:177–189
Fig. 7 Behavioral stripes for each group. Bar thicknesses express
behavioral density in terms of t-patterns encompassing each behav-
ioral element, during a 10-min observational window (x-axis). For all
this propensity is increased by DZP, it is unclear why
climbing, representing the vertical exploration of walls,
results significantly reduced (Fig. 2). A similar issue arises
for edge-sniff and head-dip: if the former shows a dose-
dependent duration increase, in agreement with a possible
drug-induced facilitation of exploratory propensity, the
latter remains unexpectedly unmodified. Thus, injected
animals present not homogeneous modifications of explor-
atory components duration which, instead, should be
influenced similarly by anti-anxiety treatment. It is possible
to conclude that exploratory activities, in their various
expressions, are quite differently affected by DZP. More-
over, a discrepancy is also present for grooming compo-
nents, whose durations demonstrate that face grooming is
significantly reduced, whereas front-paw licking and body
grooming remain unmodified (Fig. 2). These findings call
for complex underlying behavioral dynamics which cannot
183
groups, t-patterns encompassing HPL have not been detected. Data
were obtained from the analysis of 56 rats. For abbreviations, see
Table 1
be appraised by means of “conventional” descriptive
approaches.
T-patterns number and length
DZP induces, at all doses, a clear-cut reduction of t-patterns
total number. A decrease is evident between saline and
2 mg/kg diazepam-injected groups: a difference of roughly
1,000 t-patterns (par. 3.2—Table 2). Accordingly, mean
occurrences for each group demonstrate that diazepam
administration induces a dose-dependent decrease of
t-patterns (Fig. 3). Such modification can be further
appreciated from two different points of view: firstly, by
taking into consideration the mean number of t-patterns on
the basis of their length; secondly, by taking into
consideration the mean number of t-patterns on the basis
of their different composition. The number of t-patterns
184 Psychopharmacology (2011) 215:177–189

Table 2 Number and percent values of t-patterns encompassing each behavioral element in saline and diazepam-injected groups

T-patterns encompassing each Saline Diazepam

behavioral element
0.25 mg/kg 0.5 mg/kg 2 mg/kg

2,750a 2,289a 1,937a 1,762a

Wa 1,199 (43.6%) 1,192* (52.07%) 1,002* (51.72%) 1,163* (66.01%)

Cl 1,567 (56.98%) 1,112* (48.58%) 760* (39.23%) 733* (41.6%)
Re 16 (0.58%) 308* (13.45%) 0 (0%) 0 (0%)
IS 1,670 (60.72%) 1,292* (56.44%) 1,104* (56.99%) 956* (54.25%)
ES 1,036 (37.67%) 641* (28.01%) 407* (21.01%) 508* (28.83%)
HD 1,199 (43.6%) 623* (27.21%) 777* (40.11%) 879* (49.88%)
FPL 42 (1.52%) 48 (2.09%) 36 (1.85%) 10* (0.56%)
HPL 0 (0%) 0 (0%) 0 (0%) 0 (0%)
FG 79 (2.87%) 52 (2.27%) 45 (2.32%) 10* (0.56%)
BG 57 (2.07%) 47 (2.05%) 44 (2.27%) 0 (0%)
Im 46 (1.67%) 40 (1.74%) 122* (6.29%) 65* (3.68%)

Data were obtained from the analysis of 56 rats. For abbreviations, see Table 1
*p≤0.05, significant differences in comparison with saline (Chi-square test)
a
Total number of detected t-patterns for each group

encompassing three, four and five elements, and the behavioral complexity and variability. Therefore, it is
number of t-patterns of different composition results were possible to suggest that pharmacological modulation of
largely reduced by DZP administration (Figs. 4 and 5). anxiety level significantly affects the temporal structure of
These pieces of evidence demonstrate that DZP modifies behavioral response to anxiety in HB.

Fig. 8 Behavioral stripes for each group. Bar thicknesses express Data were obtained from the analysis of 56 rats. For abbreviations, see
behavioral density in terms of t-patterns encompassing specific Table 1
behavioral sequences, during a 10-min observational window (x-axis).
Psychopharmacology (2011) 215:177–189 185

Table 3 Number and percent values of t-patterns encompassing specific sequences in saline and diazepam-injected groups

T-patterns encompassing Saline Diazepam

specific sequences
0.25 mg/kg 0.5 mg/kg 2 mg/kg

2,750a 2,289a 1,937a 1,762a

Wa→Cl 481 (17.49%) 267* (11.66%) 212* (10.94%) 240* (13.62%)

Cl→Wa 180 (6.54%) 278* (12.14%) 155* (8.01%) 254* (14.41%)
ES→HD 298 (10.83%) 165* (7.2%) 87* (4.49%) 156* (8.85%)
HD→ES 109 (3.96%) 152* (6.64%) 140* (7.22%) 197* (11.18%)
ES→Wa 309 (11.23%) 127* (5.54%) 115* (5.93%) 341* (19.35%)
HD→Wa 209 (7.6%) 140* (6.11%) 193* (9.96%) 306* (17.36%)
FG→BG 57 (2.07%) 37 (1.61%) 32 (1.65%) 0 (0%)
BG→FG 0 (0%) 0 (0%) 0 (0%) 0 (0%)

Data were obtained from the analysis of 56 rats. For abbreviations, see Table 1
*p≤0.05, significant differences in comparison with saline (Chi-square test)
a
Total number of detected t-patterns for each group

Fig. 9 Within groups statistical validation. x-Axis, t-patterns length distribution in real data and in random generated data (±1 SD) assessed for
ten random runs as provided by THEME. y-Axis, number of independent t-patterns. Data were obtained from the analysis of 56 rats
186
T-patterns composition
Individual behavioral components
DZP induces clear and significant changes of t-patterns
occurrences containing single behavioral components and
specific sequences.
Several studies have demonstrated a significant increase
of locomotor activity following administration of non-
sedative doses of anti-anxiety molecules in rodents (Harada
et al. 2006; Kralic et al. 2002; Shimada et al. 1995). Our
results are well in agreement with such findings. Indeed, all
t-patterns including walking (Table 2) are significantly
increased. On the basis of the relationships between walking
and climbing, (Casarrubea et al. 2009a, b, 2010a), DZP
should influence horizontal (walking) and vertical (climbing)
exploratory patterns in a similar way. Instead, we observed
that t-patterns including climbing were significantly reduced.
Such opposite effects, evident at higher dose, were also
observed on t-patterns including edge-sniff (decrease) or
head-dip (increase), hole exploratory elements related to one
another (Casarrubea et al. 2009a, 2010a). Further on, these
data confirm that DZP heavily influences the temporal
structure of all exploratory activities.
Animal behavior in HB has been suggested to originate
from an antagonism between an exploratory and an escape
propensity (Montgomery 1955; Saitoh et al. 2006). The
results of this antagonism are strongly related to anxiety
level. Such conflictual condition may explain what has
been observed in the present study concerning non-
consensual DZP effects on t-patterns including different,
ethologically characterized, behavioral elements of rat
exploratory activity. Diazepam-induced anxiety level re-
duction decreases t-patterns encompassing climbing and
edge-sniff: this finding may be explained through a lower
need to look for an escape way (climbing) and a lower need
to explore the hole border (edge-sniff) before the animal
inserts its head inside (head-dip). On the contrary, anxiety
level reduction determines the increase of t-patterns
including walking and head-dip, provoking a greater
propensity to explore the novel environment (walking)
and the hole inside (head-dip). Further proof of DZP’s
heavy influence on behavioral structure arises from the
significant reduction of t-patterns encompassing immobile
sniffing which has been suggested to represent a behavioral
component related to cognitive processes of environmental
information analysis (Casarrubea et al. 2008, 2009a,
2010a). The significant percent reduction of t-patterns
encompassing all grooming components at 2 mg/kg dose
and the significant increase of immobility at 0.5 and 2 mg/kg
might arise from a drug-induced sedation which negatively
influences motor activities. However, a sedative effect can be
excluded for several reasons: firstly, because of the significant
Psychopharmacology (2011) 215:177–189
increase of walking duration (Fig. 2) and percent of t-patterns
encompassing walking (Table 2); secondly, because immo-
bility durations are not significantly modified following DZP
(Fig. 2); finally, because the doses of DZP, used in this
experiment, were chosen since they have been demonstrated
not to provoke locomotor activity impairments (Casarrubea
et al. 2009b; Harada et al. 2006). Since grooming activity
has been reported to be related with anxiety condition
(Kalueff and Tuohimaa 2004; Spruijt et al. 1992), diazepam-
induced anxiolysis could provoke a consensual decrease
both of anxiety episodes and of t-patterns containing
grooming behavioral elements.
Behavioral sequences
In untreated rodents, the behavioral components edge-sniff,
head-dip, walking, climbing, front-paw licking, face
grooming, and body grooming are arranged in the follow-
ing sequences: (edge-sniff ➔ head-dip), (head-dip ➔ edge-
sniff ), (walking ➔ climbing), (climbing ➔ walking), and
(front-paw licking ➔ face grooming ➔ body grooming)
(Casarrubea et al. 2010a). Present results demonstrate that
t-patterns including the sequences (edge-sniff ➔ head-dip)
and (walking ➔ climbing) are significantly reduced by DZP
at all doses (Table 3). On the contrary, t-patterns including
the sequences (head-dip ➔ edge-sniff) and (climbing ➔
walking) are significantly increased (Table 3). Since
climbing can only occur at walls and the animal, as a
result of the anxiolysis, walks more and remains prevalently
in central areas of the arena (where there are no walls), a
reduction of the t-patterns including (walking ➔ climbing)
sequences can be observed. Walking activity increase may
also explain the increase of (climbing ➔ walking) sequen-
ces. Indeed, if walking increases, transition possibilities
from climbing to walking will be consensually increased.
Such simple, phenomenological explanations give account
only in part for the changes in the temporal sequences
observed; moreover, they do not take into any consideration
the functional meaning of the described t-patterns. Indeed,
since surrounding walls and ground holes would trigger a
cognitive process of choosing (i.e., whether to explore one
of the walls or one of holes for a possible escape way),
(walking ➔ climbing), and (edge-sniff ➔ head-dip)
sequences would express the animal cautious approach to
unknown environment and, as a consequence, would be
heavily dependent on emotion and motivation: for instance,
the subject first sniffs the outside of hole to be reassured
(do-Rego et al. 2006), then the inside to continue the
environmental exploration. On the other hand, (climbing ➔
walking) and (head-dip ➔ edge-sniff) sequences could be
more indicative of exploratory activities running to comple-
tion, independently from a preliminary watchful approach: the
rat first puts its head inside hole, after that, when its head is
Psychopharmacology (2011) 215:177–189
removed from the hole, it completes exploration by sniffing
hole border. Thus, DZP would provoke an increase of those
exploratory sequences like (head-dip ➔ edge-sniff) and
(climbing ➔ walking) detached from emotional/motivational
control. Diazepam-induced decoupling between exploratory
activities and underlying emotional control could be consid-
ered in terms of animal risk assessment. Indeed, rodents are
well known to be very cautious toward unfamiliar environ-
ments and/or objects. Risk assessment is extremely important
in such mammals and, accordingly, it has been described as a
resulting equilibrium between propensity to investigate a
novel stimulus and the fear it causes (Augustsson et al. 2005).
This delicate balance might be considered strictly dependent
on emotional drives (Cardinal et al. 2002). Evidence that
diazepam-administered rats poke their head into a hole
significantly more often, without any preliminary sniffing
activity of its border, would suggest the appearance of a
disadvantageous behavior for risk assessment. This fearless
approach to exploration is consistent with observations also
carried out in monkeys with surgical lesions of limbic
system. For instance, Klüver and Bucy have reported the so-
called ‘psychic blindness’, that is, an evident fearless
behavior characterized by the ‘strong tendency to approach
animate and inanimate objects without hesitation’ (Klüver
and Bucy 1997). Moreover, in human beings, low doses of
benzodiazepines induce cognitive impairments, weakening
of judgment capacity, and serious disinhibitory behavioral
effects such as self-control loss (Trevor and Way 2006).
DZP effects on t-patterns temporal organization may
depend on benzodiazepines affinity for various GABA-A
receptors subtypes (González-Pardo et al. 2006; Möhler et
al. 2002). By means of the radioligands [3H]diazepam and
[3H]flunitrazepam, CNS benzodiazepines binding sites
have been demonstrated to be highly localized in rat limbic
system (Blaestrup and Squires 1977). Accordingly, acute IP
administrations of various GABA-A modulators, such as
diazepam, alprazolam, or zolpidem produces significant
variations in the metabolic activity of various rat limbic
regions (González-Pardo et al. 2006) that have been
suggested to be implicated in the evaluation of motivation-
ally significant inputs (Cardinal et al. 2002). Therefore, the
behavioral changes, detected in present study, may depend
on DZP activity in those limbic structures likely linked to
risk assessment, a process critically influenced by environ-
mental cues.
T-pattern analysis in psychopharmacological studies
Behavioral Pharmacology has been conservative in the
renewal of its methods. An elegant review by Kalueff et al.
(2007) underlines an unproductive and persistent associa-
tion: the lack of converging findings and new/alternative
approaches in the study of important disorders such as
187
anxiety and/or depression. As a matter of fact, the largest
amount of behavioral studies on psychopharmacology of
anxiety and depression has employed only a few descriptive
parameters (i.e., latencies, durations, and frequencies) of
distinct behavioral elements. In the hole board test, one of
the most investigated elements is the head-dip. However,
quite diverging findings surround this important component
of the behavioral response to anxiety since increases
(Takeda et al. 1998), decreases (Pellow et al. 1985), or no
modifications (Sayin et al. 1992) of head-dip frequencies
have been described following anti-anxiety drugs. Descrip-
tive/quantitative approaches such as assessments of latencies
or durations are useful because they provide information
concerning each investigated behavioral element. However,
such approaches are not able to detect the most important
features of behavior: the functional relationships among
behavioral elements. Therefore, descriptive approaches to
psychopharmacological studies should be, where possible,
partnered with different suitable techniques of behavioral
investigation.
A very important support to these researches can be
given by multivariate analyses and, in particular, by the
t-pattern one through which temporal parameters of a
behavior, not discernable/interpretable by means of quanti-
tative assessments, can be detected. Indeed, our results
suggest that complexity and/or variability of behavioral
organization and the possible temporal rearrangement of
t-patterns containing specific exploratory sequences such as
(edge-sniff ➔ head-dip), (head-dip ➔ edge-sniff), (walking ➔
climbing), and (climbing ➔ walking) may represent useful
parameters to evaluate and compare different classes of anti-
anxiety molecules.
Conclusions
The present study demonstrates, for the first time, that the
temporal structure of the Wistar rats behavioral response to
anxiety in HB, significantly changes following pharmaco-
logical manipulation of anxiety condition. Moreover, our
results suggest that multivariate t-pattern analysis of rat
behavior in HB could be a further, useful tool to study
possible effects of drugs influencing anxiety condition.
Acknowledgments This work was supported by a grant from the
“University of Palermo”, Palermo, Italy.
References
Adamec R, Head D, Blundell J, Burton P, Berton O (2006) Lasting
anxiogenic effects of feline predator stress in mice: sex differences
in vulnerability to stress and predicting severity of anxiogenic
response from the stress experience. Physiol Behav 88:12–29
188
Augustsson H, Dahlborn K, Meyerson BJ (2005) Exploration and
risk assessment in female wild house mice (Mus musculus
musculus) and two laboratory strains. Physiol Behav 84:265–
277
Blaestrup C, Squires RF (1977) Specific benzodiazepine receptors in
rat brain characterized by high-affinity [3H]diazepam binding.
Proc Natl Acad Sci 74:3805–3809
Bonasera SJ, Schenk AK, Luxenberg EJ, Tecott LH (2008) A novel
method for automatic quantification of psychostimulant-evoked
route-tracing stereotypy: application to Mus musculus. Psycho-
pharmacology (Berl) 196:591–602
Cardinal RN, Parkinson JA, Hall J, Everitt BJ (2002) Emotion and
motivation: the role of the amygdala, ventral striatum, and
prefrontal cortex. Neurosci Biobehav Rev 26:321–352
Casarrubea M, Sorbera F, Crescimanno G (2006) Effects of 7-OH-
DPAT and U 99194 on the behavioral response to hot plate test,
in rats. Physiol Behav 89:552–562
Casarrubea M, Sorbera F, Crescimanno G (2008) Multivariate
analysis of the modifications induced by an environmental
acoustic cue on rat exploratory behavior. Physiol Behav 93:
687–696
Casarrubea M, Sorbera F, Crescimanno G (2009a) Structure of rat
behavior in hole-board: I) multivariate analysis of response to
anxiety. Physiol Behav 96:176–179
Casarrubea M, Sorbera F, Crescimanno G (2009b) Structure of rat
behavior in holeboard: II) multivariate analysis of modifications
induced by diazepam. Physiol Behav 96:683–692
Casarrubea M, Sorbera F, Crescimanno G (2009c) Multivariate data
handling in the study of rat behavior: an integrated approach.
Behav Res Methods 41:772–781
Casarrubea M, Sorbera F, Magnusson M, Crescimanno G (2010a)
Temporal patterns analysis of rat behavior in hole-board. Behav
Brain Res 208:124–131
Casarrubea M, Sorbera F, Santangelo A, Crescimanno G (2010b)
Microstructure of rat behavioral response to anxiety in hole-
board. Neurosci Lett 481:82–87
Castañer M, Torrents C, Anguera MT, Dinusová M, Jonsson GK
(2009) Identifying and analyzing motor skill responses in body
movement and dance. Behav Res Methods 41:857–867
Choleris E, Thomas AW, Kavaliers M, Prato FS (2001) A detailed
ethological analysis of the mouse open field test: effects of
diazepam, chlordiazepoxide and an extremely low frequency
pulsed magnetic field. Neurosci Biobehav Rev 25:235–260
Cruz AP, Frei F, Graeff FG (1994) Ethopharmacological analysis of
rat behavior on the elevated plus-maze. Pharmacol Biochem
Behav 49:171–176
do-Rego JC, Viana AF, Le Maitre E, Deniel A, Rates SMK, Leroux-
Nicollet I, Costentin J (2006) Comparisons between anxiety tests
for selection of anxious and non anxious mice. Behav Brain Res
169:282–288
Espejo EF (1997) Structure of the mouse behaviour on the elevated
plus-maze test of anxiety. Behav Brain Res 86:105–112
File SE, Wardill AG (1975) The reliability of the hole-board
apparatus. Psychopharmacologia 44:47–51
González-Pardo H, Conejo NM, Arias JL (2006) Oxidative metabo-
lism of limbic structures after acute administration of diazepam,
alprazolam and zolpidem. Prog Neuropsychopharmacol Biol
Psychiatry 30:1020–1026
Grammer K, Kruck KB, Magnusson M (1998) The courtship dance:
patterns of nonverbal synchronization in opposite-sex encounters.
J Nonverbal Behav 22:3–29
Psychopharmacology (2011) 215:177–189
Harada K, Aota M, Inoue T, Matsuda R, Mihara T, Yamaji T et al
(2006) Anxiolytic activity of a novel potent serotonin 5-HT2C
receptor antagonist FR260010: a comparison with diazepam and
buspirone. Eur J Pharmacol 553:171–184
Hirschenhauser K, Frigerio D, Grammer K, Magnusson MS (2002)
Monthly patterns of testosterone and behavior in prospective
fathers. Horm Behav 42:172–181
Kalueff AV, Tuohimaa P (2004) Grooming analysis algorithm for
neurobehavioural stress research. Brain Res Prot 13:151–
158
Kalueff AV, Tuohimaa P (2005) The grooming analysis algorithm
discriminates between different levels of anxiety in rats: potential
utility for neurobehavioural stress research. J Neurosci Methods
143:169–177
Kalueff AV, Wheaton M, Murphy DL (2007) What’s wrong with my
mouse model? Advances and strategies in animal modeling of
anxiety and depression. Behav Brain Res 179:1–18
Kamei J, Hirose N, Oka T, Miyata S, Saitoh A, Yamada M (2007)
Effects of metylphenidate on the hyperemotional behavior in
olfactory bulbectomized mice by using the hole-board test. J
Pharmacol Sci 103:175–180
Kemp AS, Fillmore PT, Lenjavi MR, Lyon M, Chicz-Demet A,
Touchette PE, Sandman CA (2008) Temporal patterns of self-
injurious behavior correlate with stress hormone levels in the
developmentally disabled. Psychiatry Res 157:181–189
Kerepesi A, Jonsson GK, Miklósi A, Topál J, Csányi V, Magnusson
MS (2005) Detection of temporal patterns in dog-human
interaction. Behav Process 70:69–79
Klüver H, Bucy PC (1997) Preliminary analysis of functions of the
temporal lobes in monkeys. J Neuropsychiatry Clin Neurosci
9:606–620
Kralic JE, O’Buckley TK, Khisti RT, Hodge CW, Homanics GE,
Morrow AL (2002) GABA-A receptor alpha-1 subunit deletion
alters receptor subtype assembly, pharmacological and behavioral
responses to benzodiazepines and zolpidem. Neuropharmacology
43:685–694
Lino-de-Oliveira C, De Lima TCM, Carobrez AP (2005) Structure of
the rat behaviour in the forced swimming test. Behav Brain Res
158:243–250
Lyon M, Kemp AS (2004) Increased temporal patterns in choice
responding and altered cognitive processes in schizophrenia and
mania. Psychopharmacology (Berl) 172:211–219
Magnusson MS (2000) Discovering hidden time patterns in behavior:
T-patterns and their detection. Behav Res Methods Instrum
Comput 32:93–110
Magnusson MS (2004) Repeated patterns in behavior and other
biological phenomena. In: Oller DK, Griebel U (eds) Evolution
of communication systems: a comparative approach. The MIT,
Cambridge, pp 111–128
Möhler H, Fritschy JM, Rudolph U (2002) A new benzodiazepine
pharmacology. J Pharmacol Exp Ther 300:2–8
Montgomery KC (1955) The relation between fear induced by novel
stimulation and exploratory behavior. J Comp Physiol Psychol
48:254–260
Pellow S, Chopin P, File SE, Briley M (1985) Validation of open:
closed arm entries in an elevated plus-maze as a measure of
anxiety in the rat. J Neurosci Methods 14:149–167
Saitoh A, Hirose N, Yamada M, Yamada M, Nozaki C, Oka T, Kamei
J (2006) Changes in emotional behavior of mice in the hole-
board test after olfactory bulbectomy. J Pharmacol Sci 102:
377–386
Psychopharmacology (2011) 215:177–189
Sayin U, Purali N, Ozkan T, Altug T, Buyukdevrim S (1992)
Vigabatrin has an anxiolytic effect in the elevated plus-maze test
of anxiety. Pharmacol Biochem Behav 43:529–535
Shimada T, Matsumoto K, Osanai M, Matsuda H, Terasawa K,
Watanabe H (1995) The modified light/dark transition test in
mice: evaluation of classic and putative anxiolytic and anxio-
genic drugs. Gen Pharmac 26:205–210
Spruijt BM, Gispen WH (1984) Behavioral sequences as an easily
quantifiable parameter in experimental studies. Physiol Behav
32:707–710
189
Spruijt BM, van Hooff JA, Gispen WH (1992) Ethology and
neurobiology of grooming behavior. Physiol Rev 72:825–852
Takeda H, Tsuji M, Matsumiya T (1998) Changes in head-dipping
behavior in the hole-board test reflect the anxiogenic and/or
anxiolytic state in mice. Eur J Pharmacol 350:21–29
Trevor AJ, Way WL (2006) Sedative-hypnotic drugs. In: Katzung B (ed)
Basic & clinical pharmacology, 10th edn. McGraw-Hill, pp 347-362
Van Den Berg CL, Van Ree JM, Spruijt BM (1999) Sequential
analysis of juvenile isolation-induced decreased social behavior
in the adult rat. Physiol Behav 67:483–488
Copyright of Psychopharmacology is the property of Springer Science & Business Media B.V. and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.