Ejd;15oct00

Psychopharmacology (2000) 152:249–255DOI 10.1007/s002130000525 Muriel Boucart · Paul de Visme · Johan Wagemans
Effect of benzodiazepine on temporal integration in object perception Received: 9 June 1999 / Accepted: 17 May 2000 / Published online: 19 August 2000 Springer-Verlag 2000 Abstract Rationale: Though various psychometrical
Key words Benzodiazepines · Lorazepam · Diazepam ·
tests indicate that benzodiazepines affect vigilance, few Object recognition · Selective attention · Attentional studies have been conducted to assess the effect of blinkbenzodiazepines on attentional processes. Objective: Weused a RSVP (Rapid Serial Visual Presentation) proce-dure to investigate the effect of benzodiazepines on the attentional blink effect. It refers to the difficulty in de-tecting a probe following identification of a target within Benzodiazepines have been shown to impair various cog- a temporal window of 500 ms. Method: Three experi- nitive processes. The most extensively studied impairment mental groups were tested (placebo, lorazepam and diaz- is that of human memory, where, in addition, a pharmaco- epam). Sequences of 15 pictures were centrally dis- logical dissociation of the effects of two benzodiazepines played for 50 ms each. In a dual-task condition, observ- (lorazepam and diazepam) on implicit and explicit memo- ers were instructed (1) to identify the target (the single ry has been established (Danion et al. 1989a, 1989b; Sellal picture on a blue background) and (2) to detect the pres- et al. 1992; Vidhailhet et al. 1994; Legrand et al. 1995). In ence of a probe. In the single-task condition, subjects addition to its deleterious effect on memory, one benzodi- were asked to detect the probe. The serial position of the azepine (lorazepam) has recently been shown to impair vi- probe relative to the target was varied. Results: Perfor- sual information processing, especially perceptual integra- mance was equivalent for the three groups in the single- tion of local contour elements into a visual shape or task condition. In the dual-task condition, the attentional “Gestalt” (Giersch et al. 1995, 1996; Wagemans et al.
blink was increased in magnitude and duration for ben- 1998, Beckers et al., submitted). Though various psycho- zodiazepine-treated subjects, especially diazepam, than metrical tests indicate that benzodiazepines affect vigi- for placebo-treated subjects. A large number of intru- lance, increasing response times and errors in sustained at- sions (a tendency to report as target the name of a picture tention tasks (Kleinknecht and Donaldson 1975; Preston preceding the target) were observed in the benzodiaze- et al. 1989; van Leuwen et al. 1994), few studies have pine-treated groups. Conclusion: The results indicate been conducted to assess the effect of benzodiazepines on that benzodiazepines impair visual integration in the attentional processes directly. Johnson et al. (1995) report- temporal domain. This extends previous findings that ed a study in which the effect of a benzodiazepine (triazo- benzodiazepine impairs visual integration in the spatial lam) on spatial attention was investigated with a cueing domain. The results also suggest that benzodiazepine in- paradigm. Central cues (arrows) and peripheral cues (the crease time to disengage attention from a first to a sec- brightening of a box) were used to examine controlled (central cueing) and automatic (peripheral cueing) atten-tion allocation mechanisms. Subjects were instructed to indicate which of two target letters was present in a dis- Laboratoire de Neuroscience du Comportement, play composed of four letters. The authors found that the Université Lille 1, Bat. SN4, 59655 Villeneuve d’Ascq, Francee-mail: [email protected] cost in response times and accuracy for invalid cues, rela-tive to valid cues, was larger in the triazolam treatment group than in placebo-treated subjects, indicating that tri- Universtity of Copenhagen, Psychological Laboratory, Njalsgade 90, 2300 Copenhagen S, Denmark azolam affects the operation of disengagement of attentionfrom the cue to the target.
We were interested in the effect of benzodiazepines University of Leuven, Laboratory of Experimental Psychology,Tiensestraat 102, 3000 Leuven, Belgium on visual attention in the temporal domain. A method that has become widely used to examine the time course get are included during the operation of consolidation or of attentional processes is the rapid serial visual presen- transfer. These accounts predict a pattern of post-target tation (RSVP) paradigm. This technique involves the presentation of a sequence of stimuli in rapid succession A higher number of intrusions is also expected if tem- usually at the same spatial location. Subjects are asked to poral binding is impaired. For instance, Intraub (1985, report one or more pre-defined targets. In the classical 1989) used outline drawings of objects in RSVP se- paradigm, attention is engaged on a first target (e.g., the quences. Subjects were asked to identify the picture sur- single white letter in a sequence of black letters) by re- rounded by a frame. Intraub reported frequent migrations quiring its identification. At varying intervals following of the frame, i.e., subjects displayed a tendency to report the presentation of the first target, a second target (the the name of the picture preceding that surrounded by the probe) is presented for detection. Response is given at frame. She argued that the frame is erroneously linked to the end of the sequence. A result consistently found is the previous picture that is still in the buffer being pro- that subjects experience difficulties in reporting a second cessed. Given that the target was defined by the color of target if it occurs in a temporal window of about 500 ms its background in the present study, impaired temporal following the first target. This phenomenon has been binding should result in a high number of pre-target in- called the “attentional blink” effect (Raymond et al.
Two main classes of explanations have been proposed for this effect: a memory account (Chun and Potter 1995; Raymond et al. 1995; Giesbrecht and Di Lollo 1998;Vogel et al. 1998) and a dual-task interference account (Ward et al. 1996, 1997; Jolicoeur 1999). The memory The study was approved by the Ethical Committee of Alsace. account states that the operation of transferring a first Fifty-four healthy volunteers (18–35 years of age) were recruited target into visual working memory (Vogel et al. 1998) or from the Faculty of Medicine of the University Louis Pasteur in the operation of consolidation of a first target in short- Strasbourg. They were native French speakers. They had no medi- term memory (Chun and Potter 1995) to be reported at cal illness or history of alcoholism, drug abuse or tobacco con-sumption of more than ten cigarettes per day. They were not the end of the sequence monopolizes attentional resourc- chronic users of benzodiazepines and had not taken any concomi- es for some time (about 500 ms) thus reducing the pro- tant medication for at least 21 days. Subjects were instructed to cessing of subsequent stimuli due to capacity limitations abstain from beverages containing caffeine or alcohol for the 24 h (Neuman 1987). The interference account is based on the prior to the study. All subjects were tested in the morning after an idea that attention to an object for the purpose of identi- overnight fast. They all had normal or corrected-to-normal vision.
Written informed consent was obtained from all volunteers before fication produces a pattern of long lasting interference they entered the study. Subjects were paid 1000 Francs for their (called the “dwell time”, about 500 ms) on the process- ing of subsequent stimuli. Consistent with this account,neuropsychological data on neglect patients (Husain Stimuli et al. 1997) show that the blink is about 4 times longer inpatients who have difficulties to disengage their attention The stimuli were 120 colored pictures of objects taken from chil- from the first target to the probe (see also Di Pellegrino dren books (“L’imagier du Père Castor” Flammarion and “Des im- ages et des mots” Nathan). Eight semantic categories were repre-sented: four natural categories (four-legged animals, birds, fruits In our study, we tested the effect of two different ben- and vegetables) and four categories of non-living things (vehicles, zodiazepines (lorazepam and diazepam) on the magni- furniture, tools and clothes). Each semantic category contained 15 tude and the duration of the attentional blink effect. A different objects. The pictures were digitized in PCX format. The longer attentional blink can reflect either increased time mean size was 130×130 pixels in a matrix of 200×200 pixels.
of attentional disengagement (Husain et al. 1997; Ward Each of the 120 pictures was displayed as target (T1) on a bluebackground and as distractor on a grey background. The probe et al. 1997), increased time to consolidate or to transfer (T2) was an item belonging to none of the eight semantic catego- the target from short-term conceptual memory into work- ing memory (Chun and Potter 1995; Vogel et al. 1998) orimpaired temporal integration, since the target’s identity and the physical feature defining the target have to be in-tegrated. A way to dissociate between these accounts is The stimuli were centrally displayed on the color screen (14 inch) to check for intrusions. Indeed, a major problem with the of a Pentium II 166 MHz PC computer equipped with a VGA dwell time account of the attentional blink effect (Ward graphic card. Viewing distance was not fixed but subjects viewedthe display from a distance of approximately 35 cm. The screen et al. 1997) is that it neither predicts nor explains the resolution was 640×480 pixels spatially, and stimuli were dis- presence of intrusions (i.e., the tendency to report the played using 256 colors. The refresh rate of the screen was 60 Hz.
name of distractors present in the sequence instead of thetarget’s name), whereas intrusions can be anticipatedfrom both the Chun and Potter (1995) model and the re- vised version (Vogel et al. 1998) of the Raymond et al.
Subjects were randomly assigned to one of three parallel groups of (1995) model because items temporally close to the tar- 18 subjects each: a placebo (lactose 190 mg) group, a lorazepam Fig. 1 An example of a sequence in which the target (T1 on a blue
On the day before the experiment, participants were presented background) is embedded in semantically related distractors. The with an example of a picture on a blue background, the same pic- ture on a grey background and the globe on a sheet of paper. Par-ticipants were then shown the whole set of pictures on paper andasked to name each picture. This test was designed to ensure thatall pictures could be identified and given the same name. Follow- 0.038 mg/kg group and a diazepam 0.3 mg/kg group. The diaz- ing the identification task participants were given 20 trials as prac- epam and lorazepam doses were equally sedative. Drugs were ad- ministered orally using a double-blind procedure. In order to eval-uate the effects of each drug at its peak plasma concentration,which is usually attained within 60 min after oral administration for diazepam and within 120 min for lorazepam (Mandelli et al.
1978; Greenblatt et al. 1993), a double-placebo procedure was used. Subjects in the diazepam group received placebo at One subject was discarded in the placebo-treated group 7.30 a.m. and diazepam at 8.30 a.m.; subjects in the lorazepam because of a high false alarm rate on the probe (above group received lorazepam at 7.30 a.m. and placebo at 8.30 a.m.; 50%). The percentage of false alarms on the probe was subjects in the placebo group received placebo at both times. Twosubjects were tested per day, the first one starting at 9.00 a.m., the higher in benzodiazepine-treated subjects than in the pla- cebo group (diazepam: 8.3%, lorazepam: 7.1%, placebo: A fixation dot was displayed centrally for 500 ms and followed 4.3%) but the difference did not reach statistical signifi- by a sequence of 15 pictures which were centrally displayed. Each picture in the sequence was displayed for 50 ms separated by a 17-ms blank interval. Each picture appeared on a grey background Figure 2 shows the mean percentage of trials in which except one (the target, T1) which appeared on a blue background.
the probe was detected correctly for each treatment T1 appeared randomly at positions 2, 3 or 4 in the sequence. The group as a function of the serial position of the probe rel- probe T2 (the globe) appeared randomly at positions 2 (134 ms), ative to the target. An ANOVA (using SYSTAT 6.0) was 4 (251 ms), 6 (385 ms), 8 (519 ms) or 10 (653 ms) after T1. A conducted. The between-subject factor was the group de- mask composed of randomly distributed colored rectangles waspresented at the end of each sequence. The target and the distrac- termined by the treatment (placebo, diazepam, loraze- tors were semantically related. An example of a sequence is dis- pam). The within-subject factors were the task condition (dual-task versus single-task) and the serial position of Performance was compared in two conditions: a dual-task was the probe relative to the target (2, 4, 6, 8, 10).
used as the experimental condition. Participants were asked (1) toidentify the single picture displayed on a blue background, and (2)to detect the presence of a globe. In the control single-task condi-tion, participants were asked to detect the presence of a globe and to ignore the picture on a blue background. Participants were pre-sented with 240 sequences in each condition: 120 sequences con-tained a globe. There were 24 trials for each position of T2 (2, 4, A significant main effect of group was found [F(2,50)= 6, 8 and 10). Responses were given verbally. The name of the pic- 4.79, P<0.012]. Accuracy in probe detection was higher ture on a blue background and the detected presence/absence of a in the placebo group than in the benzodiazepine groups globe, were typed by the experimenter on the keyboard of the (see Fig. 2). There was also a significant main effect computer at the end of each sequence. Participants were instructed of task condition [F(1,50)=158, P<0.001]: accuracy was to give the exact name of the target picture and to respond “noth-ing” if they did not identify the picture. All subjects started with higher in the single-task than in the dual-task condition.
The main effect of probe serial position was also signifi- Fig. 2 Percentage of correct probe detections (averaged over par-
was larger for benzodiazepine-treated subjects than ticipants) as a function of treatment and serial position of the for the placebo group. A significant difference was probe for the dual-task condition (left) and the single-task condi- found between placebo and diazepam [F(1,33)=11.73, tion (right). The vertical bars represent SE P<0.002] and between placebo and lorazepam [F(1,33)=4.2, P<0.049] in the dual-task condition. However, whileaccuracy in probe detection was significantly lower inthe diazepam than in the placebo group for all target-probe lags [position 2: F(1,33)=12, P<0.001, position 4:F(1,33)=6.9, P<0.013, position 6: F(1,33)=9.14, P<0.001,position 8: F(1,33)=8.9, P<0.005, position 10: F(1,33)=6.8, P<0.014], the difference between placebo and lorazepam was significant only for serial position 2[F(1,33)=5.33, P<0.027]. The interaction between groupand serial probe position did not reach statistical signifi-cance.
The percentage of correct identification of the target washigher in the placebo group (77%) than in the diazepam(56%) and the lorazepam (59%) groups [F(2,50)=13.1, Fig. 3 Distribution of intrusions as a function of treatment and se-
P<0.001]. The response “nothing” was given in about rial position of the target (T1) in the sequence: (negative=pre-tar- 8% of the sequences. Intrusions were numerous, averag- get intrusions, positive=post-target intrusions). The vertical barsrepresent SE ing about 98% of errors. Less than 1% of the errors wereperseverations (i.e., the name of an item from the previ-ous sequence) and about 1% were misidentifications cant [F(4,200)=158, P<0.001]: accuracy increased as the (i.e., the name of objects not present in the experimental target-probe lag increased. A significant interaction was observed between task condition and probe serial posi- The distribution of intrusions is displayed in Fig. 3 as tion [F(4,200)=66.4, P<0.001]. Performance was stable a function of the treatment group. Given that the target across probe serial position in the single-task condition was randomly displayed at positions 2, 3 or 4 in se- whilst a strong impairment in probe detection was ob- quences of 15 items, there were three possible positions served in the dual-task condition: the attentional blink for pre-target intrusions and 13 possible positions for effect. Task condition also interacted significantly with post-target intrusions. In fact, no intrusions were ob- group [F(2,50)=4.4, P<0.017]. This interaction resulted served beyond position 9 following the target. As can be from a larger impairment in probe detection in the ben- seen from Fig. 3, there was a predominance of pre-target zodiazepine groups than in the placebo group in the du- intrusions, especially involving the item immediately al-task condition [F(2,50)=4.72, P<0.013] whilst perfor- preceding the target (the –1 item). The number of intru- mance was not significantly affected by drug in the sin- sions was higher in the benzodiazepine groups than in gle-task condition [F(2,50)=2.16, NS]. As can be seen the placebo group. The main effect of group was signi- from Fig. 2, the magnitude of the attentional blink effect ficant both for pre-target intrusions [F(2,50)=11.85, P<0.001] and for post-target intrusions [F(2,50)=5.2, tentional account (Ward et al. 1996, 1997), the blink re- P<0.009]. Group and serial position of intrusions inter- flects the time for focal attention, having already been acted significantly [F(2, 50)=6.1, P<0.004]. Figure 3 allocated to an initial target, to be re-allocated to a subse- shows that the number of pre-target intrusions (the item quent target. As mentioned above, a benzodiazepine (tri- immediately preceding the target) was two times larger azolam) has been found to increase the time to disengage than the number of post-target intrusions (the +1 item) attention from the location of the cue to that of the target for placebo- and diazepam-treated subjects. It was four in the spatial domain (Johnson et al. 1995). If spatial and times larger for lorazepam-treated subjects.
temporal attention share common mechanisms, then ben-zodiazepines should increase the duration of the atten-tional blink effect. That was the case in the present study but this account does not predict intrusions.
In the memory account, the attentional blink effect is The results show that benzodiazepines, especially diaz- thought to reflect the time to consolidate the target into epam, increased both magnitude and duration of the at- short-term memory (Chun and Potter 1995) or to transfer tentional blink effect. The impairment in probe detection the selected target into working memory (Vogel et al.
was less pronounced with lorazepam. A differential ef- 1998). The effect of benzodiazepines on memory pro- fect of these two benzodiazepines on cognitive processes cesses has been investigated with paradigms involving has often been reported before, although both diazepam long intervals between the phase of encoding and the time of retrieval. The paradigms classically used to as- instance, lorazepam has been found to impair both ex- sess the effect of benzodiazepines on memory are either plicit and implicit memory whilst diazepam affects only free recall tasks of a list of words or priming tasks based explicit memory (Legrand et al. 1992; Vidailhet et al.
on a study phase and a test phase occurring several min- 1994) and lorazepam is more detrimental than diazepam utes later (Sellal et al. 1992; Curran and Gorenstein on perceptual integration (Giersch et al. 1995 1996; 1993; Schifano and Curran 1994; Vidailhet et al. 1994; Wagemans et al. 1998; Beckers et al., submitted). The Legrand et al. 1995; Stewart et al. 1996; Buffet-Jerrott second main result of our experiment was the dispropor- et al. 1998). From these studies, it is difficult to formu- tionate increase in the number of pre-target intrusions late hypotheses about the effect of benzodiazepines on (the name of the item preceding the target) in the two the short-term memory systems operating in RSVP para- benzodiazepine-treated groups, and particularly for lora- digms (i.e., consolidation in short-term visual memory or transfer of selected information into working memory).
The larger blink effect for benzodiazepine-treated Indeed, to our knowledge, no pharmacological study on subjects is very unlikely the result of a non-specific se- benzodiazepines involving a priming paradigm with the dation given that performance was unaffected in the target immediately following the prime has been report-single-task control condition (see Fig. 2).
ed. Therefore, we do not know whether short-term mem- Might the larger blink effect be due to a higher sensi- ory systems are impaired by benzodiazepines. Impair- tivity to masking by benzodiazepines? Fish et al. (1983) ment in transferring or consolidating the target in short- found that diazepam enhances visual masking. In their term memory should result in misidentifications, a fail- study, sequences of letters composing a word were dis- ure to report the target, or random guesses. A large num- played successively. Each letter was displayed for 70 ms ber of errors were observed in the benzodiazepine groups and the temporal interval between letters was manipulat- but they were almost exclusively due to a tendency to re- ed (10, 100 or 200 ms). Subjects were asked to report the port an item that was present in the sequence (i.e., intru- word at the end of each sequence. The decline in perfor- sions). This result is difficult to reconcile with a hypoth- mance with decreasing interval was more pronounced esis in terms of impaired short-term memory.
for diazepam-treated subjects than for placebo and me- A predominance of pre-target intrusions has been re- thylphenidate-treated subjects. From this result the au- ported before with pictures as stimuli. Intraub (1985, thors suggested that the GABA agonist diazepam en- 1989) reported a predominance of pre-target intrusions hances type A masking characterized by intrachannel in- when subjects were asked to report the identity of a pic- hibition. However, other experimental results suggest ture surrounded by a frame in a RSVP paradigm. She ar- that the attentional blink effect is unlikely to be ex- gued that the frame and the picture dissociate because plained by masking by integration (type A masking): a they are not conceptually related. She proposed that probe immediately following the target (as +1 item) is whether the frame migrates to the preceding or to the classically detected with high accuracy (see Shapiro and following picture depends on whether subjects attend to Raymond 1994; Shapiro et al. 1997 for reviews). Perfor- the frame first or to the picture first. If they attend to the mance then drops up to 500 ms of target-probe lag, thus frame first then the frame is more likely to be linked to giving rise to the typical U-shape function (see Vogel the previous picture which identification is not achieved.
et al. 1998) that characterizes type B masking by inter- The speed of processing account can also hold for our re- ruption (Turvey 1973; Breitmeyer 1984).
sults. If color (the blue background) is detected quickly Let us consider the data in the framework of the theo- whilst the identity of the previous item is still being pro- retical accounts of the attentional blink effect. In the at- cessed, then the blue background will be integrated to azepine impairs visual integration in the spatial domain(Giersch et al. 1995, 1996; Wagemans et al. 1998; Beckers et al., submitted).
This result is consistent with previous studies show- ing increased threshold for flicker-fusion under benzodi-azepines (McNab et al. 1985; Lucki et al. 1986; Danjouet al. 1992; King and Henry 1992).
The results of the present study suggest that it is the time to dissociate different visual signals that seems tobe increased by lorazepam and diazepam. Of course, wecannot exclude that time to disengage attention from astimulus to another stimulus is also increased by benzo-diazepines but this account, by itself, cannot explain thepresence of intrusions found in this study.
Acknowledgements This study was supported by a EEC Biomed
grant (N°PL962775) to the first author and a twin grant
INSERM/MGV to the first and the third authors. The authors are
Fig. 4 Percentage of correct target identification (averaged over
grateful to Dr. Welsch and Dr. Giersch for selecting participants, participants) as a function of the serial position of the target in the to Isabelle Meyer and Tom Beckers for running subjects and to Felix Wichmann for improving the language of the final version ofthis paper.
the previous picture thus resulting in a high number ofpre-target (T1–1) intrusions. A high number of pre-targetintrusions can reflect either impaired temporal integra- tion (i.e., a longer temporal interval would be required inbenzodiazepine-treated subjects to dissociate temporally Breitmeyer B (1984) Visual masking. Oxford University Press, adjacent pictures), or it can reflect a slower decay of the Buffet-Jerrott SE, Stewart SH, Teehan MD (1998) A further exam- representations activated by pictures preceding the tar- ination of the time-dependent effects of oxazepam and loraze- get. There is evidence that temporal integration is affect- pam on implicit and explicit memory. Psychopharmacology ed by benzodiazepines. For instance, the flicker-fusion threshold is lowered by benzodiazepines. A longer inter- Chun MM, Potter MC (1995) A two-stage model for multiple tar- get detection in rapid serial visual presentation. J Exp Psychol stimulus interval is required for subjects to perceive two independent stimuli (Kleinknecht and Donaldson 1975; Curran HV, Gorenstein C (1993) Differential effects of lorazepam and oxazepam on priming. Int Clin Psychopharmacol 8:37– To dissociate the impaired temporal binding account Danion JM, Zimmerman MA, Willard-Schroder D, Grangé D, from the slow decay account, we introduced a 100-ms Welsch M, Imbs JL, Singer L (1989a) Diazepam induces a dis- interval between the target and the previous picture (the sociation between explicit and implicit memory. Psychophar- pre-target item). The same subjects were tested on 80 new trials. The duration of short-term semantic memory Danion JM, Peretti S, Grangé D, Bilik M, Imbs JL, Singer L (1989b) Effects of clorpromazine and lorazepam on explicit lasts about 300 ms (Irwin and Yeomans 1986; Tiberghien memory, repetition priming and cognitive skill learning in 1997). A 100-ms interval between the target and the healthy volunteers. Psychopharmacology 108:345–351 pre-target item added to three possible pre-target pictures Danjou P, Warot D, Hergueta T, Lacomblez L, Bouhours P, Puech was still within the range (three pictures×51 ms AJ (1992) Comparative study of the psychomotor and anti- each+2×17 ms ISI between pictures+100 interval be- stress effects of ritanserin, alprazolam and diazepam in healthysubjects: some trait anxiety-independent responses. Int Clin tween the pre-target item and the target=287 ms) of short-term semantic activation. If benzodiazepines slow Di Pellegrino G, Basso G, Frassinetti F (1997) Spatial extinction down the decline of activated representations then pre- on double asynchronous stimulation. Neuropsychologia 35: target intrusions should outlast a 100-ms interval. In con- Di Pellegrino G, Basso G, Frassinetti F (1998) Visual extinction as trast, if pre-target intrusions occurred because the key a spatio-temporal disorder of selective attention. Neuroreport feature (the feature defining the target: the blue back- ground) and the identity of the previous item are tempo- Fish HU, Groner M, Groner R, Menz C (1983) Influence of diaz- rally integrated, then a 100-ms delay between the two epam and methylphenidate on identification of rapidly pre-sented letter strings: diazepam enhances visual masking. items might reduce the number of intrusions (we as- sumed that the deficit in temporal integration would not Giersch A, Boucart M, Danion JM, Vidailhet P, Legrand F (1995) exceed 100 ms). The results displayed in Fig. 4 show Effects of lorazepam on perceptual integration of visual forms that pre-target intrusions were almost completely elimi- in healthy volunteers. Psychopharmacology 119:105–114 nated by the 100-ms interval, in line with the notion that Giersch A, Boucart M, Speeg-Schatz C, Kauffman-Muller F, Danion J.M. (1996) Lorazepam impairs perceptual integration benzodiazepine impairs visual integration in the tempo- of visual forms: a central effect. Psychopharmacology 126: ral domain. This extends previous findings that benzodi- Giesbrecht B, DiLollo V (1998) Beyond the attentional blink: vi- Raymond JE, Shapiro KL, Arnell KM (1992) Temporary suppres- sual masking by object substitution. J Exp Psychol Hum Per- sion of visual processing in an RSVP task: an attentional blink. J Exp Psychol Hum Percept Perform 18:849–860 Greenblatt DJ, Scavone JM, Harmatz JS, Engelhardt N, Shader RI Raymond JE, Shapiro KL, Arnell KM (1995) Similarity deter- (1993) Cognitive effect of β-adrenergic antagonists after mines the attentional blink. J Exp Psychol Hum Percept Per- single doses: pharmacokinetics and pharmacodynamics of propanonol, atenolol, lorazepam and placebo. Clin Pharmacol Schifano F, Curran HV (1994) Pharmacological models of memo- ry dysfunction? A comparison of the effects of scopolamine Husain M, Shapiro K, Martin J, Kennard C (1997) Abnormal tem- and lorazepam on word valence ratings, priming and recall.
poral dynamics of visual attention in spatial neglect patients.
Sellal F, Danion JM, Kauffman-Muller F, Grangé D, Imbs JL, van Intraub H (1985) Visual dissociation: an illusory conjunction of der Linden M, Singer L (1992) Differential effects of diaz- pictures and forms. J Exp Psychol Hum Percept Perform 4: epam and lorazepam on repetition priming in healthy volun- Intraub H (1989) Illusory conjunctions of forms, objects and Shapiro K, Raymond J (1994) Temporal allocation of visual atten- scenes during Rapid Serial Visual Search. J Exp Psychol Learn tion: inhibition or interference. In: Carr T, Dagenbach D (eds) Inhibitory processes in attention, memory and language. Aca- Irwin DE, Yeomans JM (1986) Sensory registration and informa- tional persistence. J Exp Psychol Hum Percept Perform 12: Shapiro KL, Arnell KM, Raymond JE (1997) The attentional Johnson DN, Weingartner HJ, Andreason P, George DT (1995) An Stewart SH, Rioux GF, Conolly JF, Dunphy SC, Teehan MD effect of triazolam on visual attention and information pro- (1996) Effects of oxazepam and lorazepam on implicit and cessing. Psychopharmacology 121:145–149 explicit memory: evidence for possible influences of time Jolicoeur P (1999) Dual-task interference and visual encoding. J Exp Psychol Hum Percept Perform 25:596–616 Tiberghien G (1997) La mémoire oubliée. Mardaga King DJ, Henri G (1992) The effect of neuroleptics on cognitive Turvey MT (1973) On peripheral and central processes in vision: and psychomotor function: a preliminary study in healthy inferences from an information processing analysis of masking volunteers. Br J Psychiatry 160:647–653 with patterned stimuli. Psycholo Rev 80:1–52 Kleinknecht RA, Donaldson D (1975) A review of the effects of Van Leuwen TH, Verbaten MN, Koelega HS, Camfferman G, diazepam on cognitive and psychomotor performance. J Nerv van der Gugten J, Slangen JL (1994) Effects of oxazepam on eye movements and performance in vigilance tasks with static Legrand F, Vidailhet P, Danion JM, Grangé D, Giersch A, Van der and dynamic stimuli. Psychopharmacology 114:109–118 Linden M, Singer L (1995) Differential effects of diazepam Vidailhet P, Danion JM, Kauffman-Muller F, Grangé D. Giersch and lorazepam on repetition priming on healthy volunteers.
A, van der Linden M, Imbs JL (1994) Lorazepam and diaz- epam effects on memory acquisition in priming tasks. Psycho- Lucki I, Rickels K, Geller AM (1986) Chronic use of benzodiaze- pines on psychomotor and cognitive test performance. Psycho- Vogel EK, Luck SJ, Shapiro KL (1998) Electrophysiological evi- dence for a postperceptual locus of suppression during the at- McNab MW, Foltz EL, Schweitzer J (1985) Evaluation of signal tentional blink. J Exp Psychol Hum Percept Perform 24: detection theory on the effects of psychotropic drugs on criti- cal flicker-fusion frequency in normal subjects. Psychophar- Wagemans J, Notebaert W, Boucart M (1998) Lorazepam but not diazepam impairs identification of pictures on the basis of spe- Mandelli M, Tognoni G, Garattini S (1978) Clinical pharmacoki- cific contour fragments. Psychopharmacology 138:326–333 netics of diazepam. Clin Pharmacokinet 3:72–91 Ward R, Duncan J, Shapiro K (1996) The slow time-course of Neumann O (1987). Beyond capacity. In: Heuer H, Sanders AF visual attention. Cognit Psychol 30:79–109 (eds) Perspectives on perception and action. Lawrence Ward R, Duncan J, Shapiro K (1997) Effects of similarity, difficul-Erlbaum, London ty, and nontarget presentation on the time course of visual at- Preston GC, Ward CE, Broks P, Trau, M, Stahl SM (1989) Effects of lorazepam on memory, attention and sedation in man: an-tagonism by Ro 15-1788. Psychopharmacology 97:222–227

Source: http://www.gestaltrevision.be/publications/Boucart%20et%20al%20PsPha%202000.pdf