Startle response and its prepulse modification in health and under different psychopathologies: Could we find any specific patterns?

Abstract

The startle response (SR) is a generalized defensive response elicited by the presentation of a sudden intense stimulus. The presentation of a less intense signal (prepulse) before the central stimulus (pulse) affects the amplitude and latency of SR differently depending on the prepulse lead interval. The most studied form of such changes is prepulse inhibition (PPI), i.e. a decrease in SR amplitude at lead intervals of 50-500 ms. Prepulse facilitation, i.e. an increase in SR amplitude, can also be observed at lead intervals of 2000-4500 ms. The PPI deficiency has been found in a variety of psychopathologies and it has been suggested that it is a transdiagnostic phenomenon. However, some data from the literature support the existence of specific different nosologies, such as neurophysiological, neurochemical and molecular mechanisms that cause PPI lowering and affect prepulse facilitation of SR. This review provides a comparative analysis of studies on SR prepulse modification in healthy subjects and different groups of patients with mental or neurological disorders. The results of such an analysis may help to define directions for further research to improve methods of early diagnosis and to improve the validity of translational models.

Key Words: Startle reaction; Prepulse modification; Neuropsychiatric disorders; Attention; Genes; Brain potentials

Core Tip: There are several reviews in the literature dedicated to phenomenon of prepulse modification of the startle response. In these manuscripts its deficiency in neuropsychiatric disorders regarded as transdiagnostic process. Here we review the evidence in favor of the specificity of certain prepulse modification changes for certain pathologies of pathology and consider the prospects for future research on this issue.

INTRODUCTION

Identifying the neurophysiological and behavioral processes that constitute mental health and illness at a systems level is one of the most important research tasks in the field of biological psychiatry, but the problem of how to apply the latest neurobiological research data to the diagnosis and treatment of mental disorders remains unresolved and requires further investigation. Obviously, the success of these studies requires a permanent analysis and synthesis of the data obtained in translational research, in order to determine the quality of the existing model, as well as the direction and objectives of further research to improve it, focusing on the most relevant neurobiological phenomena.

One of the most promising tools in translational studies of the neurobiological mechanisms of psychiatric disorders is the model of the startle response (SR) and prepulse modulation, exploiting an interspecies behavioral phenomenon to study mechanisms of sensory-motor gating at the stages of preattention and early attention, attention allocation at the early stages of stimulus processing, and sustained attention to incoming sensory information at the stage of development of orienting attention[1-3].

The SR is a reflexive whole-body response to sudden high-intensity stimuli. It includes orienting and defensive components aimed at immediately modifying current behavioral activity, stimulating sensory scanning of environmental stimuli, and increasing motor readiness for further specific defensive responses: Freezing, running, attacking, etc[4,5]. In rodent laboratory studies, the magnitude of the SR is usually estimated by the increase in pressure on the chamber floor as the hind legs extend (whole-body startle)[4,5]. In primates and humans, the blink component of SR is mainly recorded by the potential of the orbicularis oculi muscle [electromyography (EMG) response][6] or by the oculographic response (electrooculogram response)[4,7].

In primate studies, the most commonly used model is the acoustic SR (ASR), which is induced by tones of at least 80 decibels and a high rate of increase (no more than 10-12 ms to peak intensity). In addition to ASR, tactile SR is also used, which can be induced by airflow directed at the eye[8], electrical stimulation of the trigeminal nerve, or sharp touching of the glabella, the skin area above the bridge of the nose[9]. Visual stimulation is also used, but the intensity of the response in this case is quite low 2-20 μV compared to 50-200 μV for the acoustic response[6]. Although SR is a very conservative form of innate behavior, its amplitude and latency can be strongly modified by current environmental stimuli.

When the main intense stimulus (pulse) is preceded by a brief stimulus of moderate intensity (prepulse), changes in SR amplitude and latency are observed. The pattern of these changes depends on the length of the interval between the prepulse and the pulse [lead interval (LI)]. The most studied type of SR modification is prepulse inhibition (PPI), the phenomenon of a decrease in SR amplitude observed when the LI duration is within 30-500 ms. The PPI cannot be explained by changes in excitability within the reflex arc of the SR, as it is also observed under conditions of cross-modal stimulation[10,11].

The PPI is thought to reflect the mechanisms of sensorimotor gating that provide adaptive response patterns to multiple stimuli. It is worth noting that in the case of PPI of the SR, we are faced with the processing of successive stimuli that differ significantly in intensity and vital salience. A sudden high-intensity startle stimulus conveys information about a threatening situation and elicits a rapid defensive response that can inhibit ongoing brain activity, including the processing of a neutral prepulse. However, because the prepulse may contain information about a source of sudden threat, the existence of specific gating mechanisms can be assumed to protect the processing of the prepulse from interruption by the sensory and motor components of the SR until the processing of the prepulse and assessment of its salience are complete. The existence of ‘interruptive’ and ‘protective’ mechanisms was originally proposed by Graham[1] and subsequently confirmed by several investigators[12,13]. At the same time, optimal behavior in the defensive situation also requires a sufficient level of response to the high-intensity startle stimulus. Neuroimaging studies in rodents[14-16] have revealed some functional brain networks involved in the PPI regulation (Figure 1).

Figure 1 Brain structures involved in the mechanisms of the acoustic startle response and its prepulse inhibition. Black bar: Main intensive stimulus (pulse); Grey bar: Preceding weak stimulus (prepulse); Triangle: Amplitude of startle response; White wide arrow: Protection of prepulse processing; Black wide arrow: Protection of main stimulus (pulse) processing. Data obtained in humans and rodents using lesion, stimulation and neuroimaging techniques are summarized[4,5,14-17,45,48]. PPI: Prepulse inhibition; ASR: Acoustic startle response.

The first is the main arc of the SR, which includes the cochlear nuclei, the ventrolateral tegmental nucleus and the caudal pontine reticular nucleus. The activity of this main ASR arc is under the control of midbrain structures including the inferior and superior colliculi and the dorsal periaqueductal grey. The upstream brain structures form two modulatory networks, including those involved in prepulse protection, suppression of the SR and increase in the PPI (‘PPI network’), and those involved in facilitation of startle processing and reduction of PPI (‘ASR network’). The coordinated activity of these systems ensures the distribution of processing resources in a situation where stimuli of different salience are rapidly presented. Interestingly, the ‘ASR network’ predominantly involves the left parts of the cortex (frontal and prelimbic), whereas the ‘PPI network’ involves the right prelimbic cortex. Human neuroimaging studies have also revealed interhemispheric asymmetry when assessing the involvement of forebrain structures in the PPI regulation (see for a review, although this issue needs further investigation)[17].

It should be noted that within the 30-500 ms time interval in which PPI occurs, it is possible to distinguish a segment of 30-120 ms, which is the interval of pre-attentive processing of stimuli by higher brain structures, and a period of 120-500 ms, when signal processing occurs with the involvement of early attentional mechanisms[18,19].

Another much less studied type of ASR prepulse modification is prepulse facilitation (PPF), which is an increase in response amplitude observed at LI values of 2000-6000 ms[20-22]. This “long-lead” PPF reflects mechanisms of general alertness, attentional orientation, and attentional sustainability[20-22]. Neuroimaging studies have demonstrated that the PPI and PPF operate in both separate and shared networks[23]. In contrast to the PPI, which provides gating and resource allocation during the processing of sensorimotor information, PPF is the result of sensorimotor integration, i.e., the summation of the response to the prepulse with the response to the pulse.

The PPI deficiency has been reported in patients with schizophrenia spectrum disorders by many authors[2,10,24-41] and in the Consortium on the Genetics of Schizophrenia multi-site study[42].

The PPF has been investigated to a much lesser extent, but its deficiency in patients with schizophrenia has been demonstrated in several studies[28-30,38,40,43].

In addition to schizophrenia spectrum disorders, the PPI deficiency has been observed in a variety of psychopathologies, and it has been proposed to be considered as a transdiagnostic process in the paradigm of the dimensional approach to neuropsychiatric disorders[44]. A decrease in the level of sensorimotor regulation can also be observed in clinically healthy people without psychiatric and neurological problems[45].

At the same time, many structures have been shown to influence the PPI and PPF modulation, including the frontal cortex, amygdala, hippocampus, cerebellar cortico-striato-pallido-thalamic and frontostriatal circuits[16,17,46-48]. The mechanisms of aetiology and pathogenesis of various psychopathologies are associated with specific and rather selective changes in the activity of these structures. Thus, we can try to identify some sets of changes in PPI and PPF parameters, which are quite specific for different neurological and psychiatric pathologies.

Given the lateralization of brain structures involved in the mediation and modulation of the PPI and PPF[16,17,45-48], the laterality of the observed changes in startle modulation by prepulses in different groups of patients can be considered as a measure with potential specificity. Braff et al[2] was the first to point out the utility of recording the blink component of the ASR bilaterally. Although some authors[38-40,49-53] have reported bilateral registration of the ASR, many of them have averaged the data from both eyes when comparing experimental groups. Some studies, however, allow comparison of data from the right and left eyes[38-40,49-51].

The analysis of the effects of the length of the LI in the PPI paradigm may also be promising, as the PPI at different time points may reflect either fully automatic gating mechanisms or those dependent on early attentional processes. In fact, the pattern of activation of brain structures during the PPI test has been shown to be dependent on the length of the LI[47,48].

There may also be specific alterations in the effects of selective attention to the prepulse, which have been shown to differ between schizophrenia patients and healthy individuals[52-60], that may also show specific changes in other disorders.

Furthermore, the specificity of the PPI changes in different neurological and psychiatric disorders may be determined by the activity levels of different neurotransmitter systems and polymorphic variants of genes controlling neurotransmission processes.

These aspects deserve to be analyzed, as understanding the mechanisms of neurological and psychiatric disorders, identifying potential biomarkers and developing therapeutic approaches may be facilitated by the detection of specific patterns of alterations in sensorimotor gating and integration under different pathological conditions.

Most studies of SR prepulse modification use protocols that involve acoustic (rather than tactile) stimulation and recording of the EMG response from the orbicularis oculi muscle. Such ‘standard’ protocols also use the so-called ‘uninstructed’ paradigm, which does not include additional techniques to increase voluntary attention to the stimuli.

SEARCH STRATEGY

The search was conducted in the following electronic databases: PubMed, Web of Science and the Russian database Elibrary. The database queries were not limited to a time period. All studies were retrieved from the beginning of the database until February 2025.

The main query field contained two keywords simultaneously: Startle and prepulse. In combination with the main query, the name of one of the target neuropsychiatric disorders was included, such as schizophrenia spectrum disorder, anxiety, posttraumatic stress disorder, panic disorder, Alzheimer’s disease, mild cognitive impairment, Parkinson’s disease, depression, bipolar disorder (BD), obsessive-compulsive disorder, Tourette’s syndrome, autism spectrum disorder, neurodegenerative disorders. The restriction ‘human’ was introduced in all cases. In addition, searches were conducted for queries containing combinations of the keywords startle, prepulse and neural circuit; startle, prepulse and electroencephalogram (EEG); startle, prepulse and genes.

ASR PREPULSE MODIFICATION IN PSYCHOPATHOLOGIES: DATA OBTAINED USING ‘STANDARD’ PROTOCOLS

In studies using such ‘standard’ protocols, reduced PPI levels have been observed in obsessive-compulsive disorder, BD, autism, anxiety, post-traumatic stress disorder (PTSD), Huntington’s disease, and Tourette’s syndrome[44,45]. The data obtained in the studies of these neuroses are presented in Tables 1, 2, 3, 4, 5, and 6. The results obtained in patients with different stages of neurodegenerative disorders are also presented in Table 7. Particular attention is paid to the laterality of the effect and its dependence on the length of the LI.

Table 1 Summary of studies on prepulse inhibition and prepulse facilitation of the startle response in the cohorts of participants with schizophrenia spectrum disorders, mean ± SD.

Ref.	Groups (group size)	Mean age	Side of recording	Lead interval, ms	PPI	PPF
Xue et al[25]	Unmedicated Sch (26)/medicated Sch (20)/Contr (31)	25.4 ± 3.1	Right	30, 60, 120	Sch < Contr at LI 30 and LI 120	NA
Hammer et al[26]	Unmedicated Sch (27)/Contr (38)	27.3 ± 3.6	Right	120	Sch < Contr	NA
Mackeprang et al[27]	Unmedicated Sch (20)/Contr (20)	27.0 ± 2.4	Right	30, 60, 120	Sch < Contr at LI 30, 60, 120	NA
Ludewig et al[28]	Unmedicated Sch (24)/Contr (20)	23.6 ± 4.2	Right	30, 60, 120, 240, 2000	Sch < Contr at LI 60	Sch < Contr
Ludewig et al[29]	Medicated Sch (67)/Contr (44)	38 ± 10.4	Right	30, 60, 120, 240, 2000	Sch < Contr at LI 60	Sch < Contr
Ludewig and Vollenweider[30]	Medicated Sch (19)/Contr (24). Tested 3 times with 1-month interval	Sch 42 ± 9; Contr 30 ± 10	Right	30, 60, 120, 240, 2000	Sch < Contr at LI 60	Sch < Contr, only in the first block (out of three in total)
Düring et al[31]	Unmedicated Sch (52)/Contr (47)	25.4 ± 4.2	Right	60, 120	Sch < Contr at LI 60 and 120	NA
Csomor et al[32]	Unmedicated Sch (14)/Contr (46)	26.1 ± 1.0	Right	60, 120	Sch < Contr at LI 60	NA
Aggernaes et al[33]	Unmedicated Sch (31)/Contr (31)	25.6 ± 0.8	Right	60, 120	Sch < Contr at LI 60	NA
Hammer et al[34]	Unmedicated Sch (13)/Contr (17)	29.3 ± 5.4	Right	30, 60, 120	Sch < Contr at LI 60	NA
Bo et al[35]	Sch (31)/High risk of psychosis/Contr (31)	23.9 ± 2.1	Right	60, 120	Sch < Contr at LI 60	NA
Hedberg et al[36]	Unmedicated Sch (49)/Contr (35)	27.8 ± 0.9	Right	30, 60, 120	Sch < Contr at LI 60	NA
Wang et al[37]	Unmedicated Sch (35)/Contr (35)	20.4 ± 4.5	Right	30, 60, 120	Sch < Contr at LI 60	NA
Storozheva et al[38]	Unmedicated Sch (26 men)/Contr (25 men)	25-50	Right and left	60, 120, 2500	Sch < Contr at LI 60 (right and left)	Sch < Contr, only in the first of two PPI blocks, at the left eye
Storozheva et al[39]	Unmedicated Sch (35 men)/Contr (26 men)	35.3 ± 10.5; 26.4 ± 5.7	Right and left	60, 120, 2500	Sch < Contr at LI 60, 120 (right and left)	No difference
Kirenskaya et al[40]	Unmedicated violent Sch (44)/non violent Sch (27)/Contr (48)	30.2 ± 6.93; 29.4 ± 7.0; 27.6 ± 3.6	Right and left	60, 120, 2500	Violent Sch < contr at LI 60 ms, bilateral; Non violent Sch< at LI 60 ms, right eye	In violent Sch; No difference from Contr; Non violent Sch < Contr; Non violent < violent Sch at both eyes, in 1st PPI block
Cadenhead et al[41]	Medicated Sch (23)/Sch relatives (34)/shizotypal (11)/Contr (25)	34.4 ± 8.5; 46.1 ± 16.1; 44.4 ± 9.0; 41.1 ± 16.7	Right and left	30, 120	Sch < Contr; Sch at LI 30 ms, right eye relatives < Contr at LI 30 ms bilaterally; Shizotypal < Contr at LI 30 ms bilaterally	NA
Wynn et al[43]	Medicated Sch (90)/Sch siblings (48)/Contr (47)	42.70 ± 8.46; 38.23 ± 9.61; 35.64 ± 7.96	Left	120, 4500	No difference	Sch < Contr; Siblings < Contr

Sch: Patients with schizophrenia; Contr: Healthy control; NA: Not applicable; PPI: Prepulse inhibition; PPF: Prepulse facilitation.

Table 2 Summary of studies on prepulse inhibition and prepulse facilitation of the startle response in the cohorts of participants with mood disorders, mean ± SD.

Ref.	Groups (group size)	Mean age	Side of recording	Lead interval, ms	PPI	PPF
Matsuo et al[63]	Non-manic BD (63 depressed + 43 euthymic)/ Contr (232)	39.4 ± 8.3	Left	60, 120	BD depressed; Men < Contr at LI 120 ms	NA
Perry et al[64]	Psychotic BD (15)/Sch (16), Contr (17)	34.4 ± 9.7	Right	30, 60, 120	Sch < Contr; BD < Contr at LI 60 ms and 120 ms	NA
Perry et al[65]	MDD (15)/Contr (17)	34.1 ± 8.3	Right	30, 60, 120	No difference	NA
Quednow et al[66]	MDD (20), Contr (18)	35.4 ± 8.3	Right	150	No difference	NA
Barrett et al[67]	Euthymic BD (23)/Contr (20)	44.4 ± 10.2	Left	60, 120	No difference	NA
Rich et al[68]	Children with BD (16)/Contr (17)	13.1 ± 2.3	Right	60, 120	No difference	NA
Carroll et al[69]	BD (32)/Contr (35)	34.5 ± 7.9	Right	60, 120	No difference	NA
Giakoumaki et al[70]	BD (21) BD/siblings (21)/Contr (19)	18-50 (range)	Right	60, 120	BD < Contr; BD siblings < Contr at LI 60 ms and 120 ms	NA
Gogos et al[71]	BD (31)/Contr (35)	21-61 (range)	Right	60, 120	Men BD < Contr at LI 60 ms; Women BD > Contr at LI 120 ms	NA
Ivleva et al[72]	BD (38)/Sch (62) relatives (32 Sch and 29 BD)/Contr (53)	40.58 ± 10.73; 36.18 ± 10.37; 41.97 ± 10.84; 32.19 ± 14.98; 36.84 ± 11.35	Right	120, 4500	No difference	No difference
Sánchez-Morla et al[73]	BD (52)/Contr (50)	70.9 ± 8.3	Right	60, 120	BD < Contr at LI 60 ms and 120 ms	NA
Matsuo et al[74]	MDD (221)/Contr (250)	18-64 (range)	Left	60, 120	Men MDD < Contr at LI 120 ms; Women no difference	NA
Massa et al[75]	Sch (143), Sch-Fam (178), SAD (123), SAD-Fam (152), BD (138), BD-Fam(183), Contr (226)	37.87 ± 12.77; 35.49 ± 13.48; 40.17 ± 15.89; 34.13 ± 11.93; 43.47 ± 15.87; 38.28 ± 12.17; 39.43 ± 16.28	Right	120, 4500	No difference	No difference

Sch: Patients with schizophrenia; Contr: Healthy control; NA: Not applicable; PPI: Prepulse inhibition; PPF: Prepulse facilitation; BD: Bipolar disorder; MDD: Major depressive disorder; SAD: Schizoaffective disorder; Sch-Fam: Relatives of patients with schizophrenia; BD-Fam: Relatives of patients with bipolar disorder.

Table 3 Summary of studies on prepulse inhibition and prepulse facilitation of the startle response in cohorts of participants with trauma- and anxiety-related disorders, mean ± SD.

Ref.	Groups (group size)	Mean age	Side of recording	Lead interval, ms	PPI	PPF
Post-traumatic stress disorder
Grillon et al[76]	Unmedicated PTSD veterans (21)/non PTSD veterans) (11)/non-PTSD civilians (17)	41.3 ± 5.0	Left	120	PTSD < non-PTSD civilians	NA
Grillon et al[77]	Unmedicated PTSD veterans (34)/non PTSD veterans) (17)/non-PTSD civilians (14)	44.1 ± 3.8	Left	120	PTSD < non-PTSD civilians	NA
Grillon et al[78]	Unmedicated PTSD veterans (21)/non PTSD veterans (15)/non-PTSD civilians (22)	47.6 ± 3.6	Left	120	PTSD < non-PTSD civilians	NA
Butler et al[79]	No data on medication PTSD veterans (20)/non PTSD veterans (18)	40.9 ± 4.4	Right	120	No difference	NA
Ornitz et al[80]	Unmedicated PTSD children (6)/control children (6)	8-13 (range)	Left	120, 250, 800, 2000	PTSD < control (at LI 120)	PTSD > control
Lipschitz et al[81]	PTSD girls (28)/control girls (23)	16.5 ± 2.2	Left	120, 2000	No difference	No difference
Holstein et al[82]	PTSD patients (27)/control (24)	38.6 ± 2.8	Right	60, 120, 2000	No difference	No difference
Pineles et al[83]	PTSD women (22)/trauma control women (25)	31.98 (9.7)	Left	120	PTSD < control	NA
Meteran et al[84]	Refugees with PTSD (25)/refugees without PTSD (20)	45.8 ± 7.3	Right	60, 120	No difference	NA
Acheson et al[85]	Combatants with PTSD (46)/combatants without PTSD (1182)	22.7 ± 1.6	Left	30, 60, 120	PTSD < combatants without PTSD, at LI 30 ms and 60 ms	NA
Echiverri-Cohen et al[86]	PTSD (28)/trauma without PTSD (27)/control (19)	32.3 ± 5.8	Left	30, 60, 120	PTSD < trauma without PTSD and control at LI 30 ms and 60 ms	NA
Panic disorders
Ludewig et al[87]	Panic disorder (20)/control (21)	31.5 ± 10.0	Right	30, 60, 120, 240 2000	Panic disorder < control at LI 30, 60, 120, 240 ms	No difference
Ludewig et al[88]	Panic disorder unmedicated (14)/panic disorder medicated (24)/control (28)	35 ± 8.1	Right	30, 60, 120, 240 2000	Panic disorder < control at LI 30, 60, 120, 240 ms	No difference
Larsen et al[89]	Panic disorder (12)/social phobia (22)/control (15)	33.6 ± 9.2	Right	100	No difference	NA
Storozheva et al[50]	General anxiety disorder + panic disorder (69)/control (50)	31.8 ± 0.9	Left and right	60, 120 2500	No difference	Patients < control
Anxiety traits
Duley et al[49]	High anxiety trait (15)/low (23) anxiety trait	21.2 ± 0.24	Left and right	30, 60, 120	High < low at 30, 60, 120 ms averaged from both eyes	NA
De Pascalis et al[90]	High anxiety trait (15)/low (23) anxiety trait	24.6 ± 3.8	Left	30, 60, 120	No difference (higher amplitude in high anxiety group)	NA
McMillan et al[91]	High (25)/low (25) anxiety sensitivity trait	22.8 ± 3.1	Left	120	High < low	NA
Nocturnal enuresis
Ornitz et al[92]	ADHD children (30)/NE children (13)/NE + ADHD children (17)/control children (42)	6-11 (range)	Left	60, 120, 4000 (control)	NE < non NE + ADHD, NE + ADHD < control (at LI 120 ms)	No difference

NA: Not applicable; PPI: Prepulse inhibition; PPF: Prepulse facilitation; PTSD: Post-traumatic stress disorder; ADHD: Attention deficit hyperactivity disorder; NE: Nocturnal enuresis.

Table 4 Summary of studies on prepulse inhibition and prepulse facilitation in the cohorts of participants with obsessive-compulsive disorder, mean ± SD.

Ref.	Groups (group size)	Mean age	Side of recording	Lead interval, ms	PPI	PPF
Hoenig et al[93]	OCD (34)/Contr (33)	31.5 ± 1.6	Right	120	OCD < Contr	NA
Swerdlow et al[94]	OCD (34)/Contr (33)	36.2 ± 2.1	Right	100	OCD < Contr	NA
de Leeuw et al[95]	Unmedicated OCD (25)/Contr (25)	32 ± 7.4	Right	100	No difference	NA
Ahmari et al[96]	Unmedicated OCD (22)/Contr (22)	31 ± 6.3	Right	100	OCD < Contr	NA
Schleyken et al[97]	Treatment refractory OCD (15)/Contr (21)	20-64 (range)	Right	60, 120, 200	No difference	NA
Steinman et al[98]	Unmedicated OCD (45)/Contr (62)	28.2 ± 4.3	Right	120	OCD < Contr (in female only)	NA

PPI: Prepulse inhibition; PPF: Prepulse facilitation; Contr: Healthy control; NA: Not applicable; OCD: Obsessive-compulsive disorder.

Table 5 Summary of studies on prepulse inhibition and prepulse facilitation of the startle response in the cohorts of participants with Tourette syndrome, mean ± SD.

Ref.	Groups (group size)	Mean age	Side of recording	Lead interval, ms	PPI	PPF
Castellanos et al[52]	ADHD (7)/ADHD + TS (7)/Contr (14)	11.4 ± 1.7	Left and right	30, 60, 120, 250 (tactile pulse and prepulse)	ADHD + TS < Contr ADHD + TS < ADHD across all LI	NA
Swerdlow et al[53]	TS (10)/Contr (14)	9-17 (range)	Left and right	120 (tactile pulse and prepulse)	TS < Contr	NA
Schleyken et al[97]	Treatment refractory TS (6)/Contr (21)	20-64 (range)	Right	60, 120, 200 (80 dB)	No difference	NA
Buse et al[99]	TS (22)/Contr (22)	13.5 ± 1.62	Left	140 (tactile pulse and prepulse)	TS < Contr	NA
Zebardast et al[100]	TS (17)/Contr (16)	30.4 ± 8.9	Left	100 (tactile pulse and prepulse)	No difference	NA

PPI: Prepulse inhibition; PPF: Prepulse facilitation; Contr: Healthy control; NA: Not applicable; TS: Tourette syndrome; ADHD: Attention deficit hyperactivity disorder.

Table 6 Summary of studies on prepulse inhibition and prepulse facilitation of the startle response in cohorts of participants with autistic spectrum disorders, mean ± SD.

Ref.	Groups (group size)	Mean age	Side of recording	Lead interval, ms	PPI	PPF
McAlonan et al[101]	Autistic (12)/Contr (14)	32.5 ± 8.2	Right	30, 60, 120	Autistic < Contr at LI 120 ms	NA
Perry et al[102]	Autistic (14)/Contr (16)	28.7 ± 5.4	Right	30, 60, 120	Autistic < Contr at LI 60 ms	NA
Yuhas et al[103]	FXS - A (17)/FXS + A (15)/IA (15)/Contr (18)	16.4 ± 1.8	Not specified	60, 120, 240	FXS - A < Contr FXS + A < Contr IA vs Contr; no diffference	NA
Sinclair et al[104]	MCDD (11)/autistic (12)/Contr (11)	11.4 ± 1.3; 11.8 ± 1.7; 11.1 ± 1.9	Not specified	120	No difference	NA
Kohl et al[105]	Autistic (17)/Contr (17)	41.8 ± 8.1	Right	60, 120, 240	No difference	NA
Madsen et al[106]	Autistic (18)/Contr (34)	11.2 ± 0.8	Right	60, 120	Autistic > Contr at LI 120 ms	NA
Takahashi et al[107]	Autistic (17)/Contr (27)	11.3 ± 1.2	Right	120	No difference in PPI, ASR amplitude; autistic > Contr at low pulse intensity	NA

PPI: Prepulse inhibition; PPF: Prepulse facilitation; Contr: Healthy control; NA: Not applicable; MCDD: Multiple complex developmental disorders; IA: Idiopathic autism; FXS: Fragile X syndrome; A: Autism.

Table 7 Summary of studies on prepulse inhibition and prepulse facilitation of the startle response in cohorts of participants with neurodegenerative disorders, mean ± SD.

Ref.	Groups (group size)	Mean age	Side of recording	Lead interval, ms	PPI	PPF
Aygün et al[108]	REM sleep disorders (19)/Contr (17)	43.8 ± 7.60	Left	200	REM sleep disorders < Contr	NA
Zoetmulder et al[109]	REM sleep disorders (12)/PD (40)/MSA (10)/Contr (20)	56.3 ± 13.0; 62.7 ± 6.6; 63.4 ± 6.7; 56.4 ± 10.0	Right	30, 60, 120, 300	MSA < Contr at 60 ms and 120 ms	NA
Hanzlíková et al[110]	Functiomal movement disorders (22)/Contr (22)	47.5 ± 9.8	Left and right	100 (electrical stimulation)	Functional movement disorders < Contr	NA
Millian-Morell et al[111]	PD (54)/Contr (35)	70.4 ± 9.8	Right	60, 120, 1000	PD > Contr at LI 120 ms	PD > Contr
Aziz[112]	Mild AD (20)/first degree (20)/Contr (30)	79.4 ± 4.7	Right	30, 60, 120, 2000	Mild AD < Contr/first degree < Contr (trend) at LI 120 ms	No difference
Ueki et al[113]	Mild AD (20)/first degree relatives (20)/Contr (30)	70.5 ± 5.7	Right	50	Mild AD < Contr/first degree relatives > Contr	NA
Salem et al[114]	AD + MCI (22)/Contr (33)	72.6 ± 4.3	Right	30, 60, 120	No difference	NA
Hejl et al[115]	AD (26)/MCI (22)/Contr (49)	73.1 ± 3.5	Right	30, 60, 120	No difference	NA
Swerdlow et al[116]	HD (22)/Contr (22)	45.2 ± 6.4	Right	30, 60, 120 acoustic and tactile	HD < Contr across al LI	NA

PPI: Prepulse inhibition; PPF: Prepulse facilitation; Contr: Healthy control; PD: Parkinson disease; REM: Rapid eye movements; MSA: multiple system atrophy; AD: Alzheimer’s disease; MCI: Mild cognitive impairment; HD: Huntington disease.

Schizophrenia

The most studied nosology is schizophrenia, and the results of some investigations in this cohort are shown in Table 1. Data on the effect of therapy on prepulse changes in SR are conflicting[24,25,33,61], so we tried to provide data from patients who were never medicated or who had discontinued medication. The exceptions were the studies that assessed PPF and those that assessed SR bilaterally.

The results of 18 studies are shown in Table 1. The protocols of all studies presented included assessment of the PPI at LI 120 ms, and a PPI deficit was observed in patients with schizophrenia in 5 studies (28%)[25-27,31,39]. Of the 15 studies in which the PPI was tested at LI 60 ms[25,27-40,43]. 14 (93%) showed a deficit in patients with schizophrenia. In one study[25], the PPI decrease in patients did not reach the level of statistical significance, although according to the figures it could be considered at the trend level. The PPI at LI 30 ms was tested in 9 studies[25,27-30,34,36,37,41] and its decrease in patients was found in 3 studies (33%). These results suggest that the most pronounced PPI deficit in schizophrenia is observed at LI 60 ms, which is consistent with the results of the Consortium on the Genetics of Schizophrenia multicentre study[42] (but see also the meta-analysis by San-Martin et al)[62].

The fact that the vast majority of researchers have recorded EMG from the orbicularis oculi of the right eye complicates the analysis of another PPI parameter, the laterality of the observed deficit.

In the studies conducted by our team using bilateral registration in patients weaned off medication, a reduction in the PPI was observed in both eyes[38-40]. However, it should be noted that these studies included patients who had committed crimes. A crime-specific analysis showed that a reduction in the PPI in the left eye was only observed in patients with a history of aggressive crime (homicide or aggravated assault), whereas right-sided PPI deficits showed no association with the characteristics of the crime[40]. Cadenhead et al[41] also used bilateral recording and analyzed data from the right and left eyes separately at 30 and 120 ms LI. A decrease in the PPI was found in the group of patients at LI 30 ms, which was more pronounced in the right than in the left eye. In a study where only the left eye was recorded, no PPI differences were found between controls and patients; however, it should be noted that the length of the LI was 120 ms[63].

Of the 7 studies that assessed PPF[28-30,38-40,43], 6 found a deficit in facilitation in schizophrenic patients. The decrease in the PPF was mostly observed in the first blocks of the test session, when the conditioned reflex for that time had not yet developed. There was no clear lateralization of the PPF deficit.

Mood disorders

The results of the studies in patients with mood disorders are shown in Table 2. It includes data from patients with BD, major depressive disorder, schizoaffective disorder (SAD), and relatives of patients with BD or SAD[63-75].

All of these studies included treated patients, as no data are available in the literature on studies of untreated or withdrawn patients with mood disorders. Three studies[64,72,75] that also included patients with schizophrenia were of particular interest, as they tested across nosologies under similar experimental conditions. Unfortunately, two of these studies found no differences between experimental groups and controls in either PPI at LI 120 ms or PPF at LI 4500 ms[72,75]. In contrast, Perry et al[64] found reduced PPI at LI 60 ms and 120 ms in both schizophrenia and BD patients.

Overall, 6 out of 13 studies (46%) found a PPI deficit in patients. In all of these studies, a decrease in the PPI was found at LI 120[63,64,70,71,73,74] and in 4 of these studies it was also found at LI 60 ms[64,70,71,73]. Thus, in contrast to schizophrenia, it cannot be concluded that there is a predominant deficit in pre-attentive gating in patients with mood disorders.

With regard to lateralization, it should be noted that in the studies with depressed patients, the PPI deficit was only observed in males at LI 120 ms and in those experiments where the side of registration was left but not right[74].

Only 2 studies assessed the PPF[72,75], showing no differences between controls and mood disorder patients.

In general, PPI deficits in mood disorders are less marked than in schizophrenia and show a significant dependence on the phase of the illness as well as on the gender of the subjects, so that females with BD may even show an increase in the PPI compared to controls.

Trauma- and anxiety-related disorders

Among trauma and anxiety disorders, PTSD is the most studied[76-86]. Notably, the majority of studies on PTSD (8 out of 11) recorded EMG from the left eye (Table 3), and differences in the PPI between patients and controls were mainly (7 out of 8 studies) observed in conditions where the left eye was assessed[76-78,80,83,84,86].

Due to the large variability in the LI sets used in the studies with PTSD cohorts, it is difficult to determine the dependence of the observed PPI changes in PTSD on the length of the LI. It should also be noted that most studies of PTSD in military personnel have found significant differences at a LI length of 120 ms when compared to civilians without PTSD, but not when compared to military personnel without PTSD[76-78]. A study by Acheson et al[85] using LI 30, 60, and 120 ms found a reduced PPI in military personnel with PTSD compared to military personnel without PTSD, and this deficit was observed at LI 30 and 60, but not 120 ms[85].

Very similar results were found by Echiverri-Cohen et al[86], who recruited an additional control group of individuals with trauma experience but without PTSD, and used LIs of 30, 60, and 120 ms. As in the Acheson et al’s study[85], PPI deficits were observed in individuals with PTSD at LIs of 30 and 60, but not 120 ms[86]. The observed phenomenon of a left-sided decrease in the PPI at short intervals is of interest for further studies of the mechanisms of PTSD and the diagnostic validity of the test.

There is a paucity of data on changes in PPF in PTSD; it is cautious to assume that it is not deficient in this disorder. There is also a paucity of research on the prepulse modification of ASR in other types of anxiety disorders[87-92]. Their results suggest impairments in sensorimotor gating and attentional allocation, which may be observed when registering from the left eye, but they do not allow conclusions to be drawn about the effects of interval length.

Obsessive-compulsive disorders

The results of the studies of prepulse modification of ASR in patients with obsessive-compulsive disorder[93-98] do not allow us to assess the effects of laterality of registration and length of LI (Table 4). In all the studies, registration was performed from the right eye and (with one exception) with LI lengths of 100 or 120 ms. In 3 of the 6 studies[93,94,96], a PPI decrease was found in patients of both sexes, and in one, the most representative study by Steinman et al[98], the deficit was found only in females. Taken together, the results of these studies suggest that patients with obsessive-compulsive disorder have a reduced PPI when recorded from the right eye and in the 100-120 ms time interval (early attentional stage).

Tourette syndrome

Deficits in Tourette syndrome were observed only in children and adolescents[52,53,97,99,100], and not in adult participants (Table 5). It should be noted that tactile stimulation was used in these studies. The SR was recorded from the left eye or bilaterally with response averaging. From the data obtained, it can be concluded that in children and adolescents with TS, a PPI decrease can be observed predominantly from the left eye with LI in the range of 100-140 ms.

Autistic spectrum disorders

Only 2[101,102] of the 7 studies[101-107] found PPI deficits in individuals with autism (Table 6). One study found reduced PPI in patients with fragile X syndrome, but not in individuals with autism[103]. Remarkably, two studies found increased PPI in children with autism[106,107]. Thus, we cannot be certain about PPI decrease in autistic spectrum disorders, and the question of the pattern of its changes in this nosology needs further study.

Neurodegenerative disorders

We analyzed data on changes in the PPI in the most common neurodegenerative disorders[108-116], such as Alzheimer’s disease and synucleinopathies (Parkinson’s disease, Lewy body dementia, Multiple System Atrophy), as well as in preclinical conditions which may represent a prodrome for the development of neurodegeneration, such as rapid eye movement (REM) sleep disorders and functional movement disorders (Table 7).

Studies of PPI in patients with preclinical signs of possible synucleinopathies have shown a PPI decrease at LI of 60 ms and in the range 100-200 ms[108-110]. Paradoxically, the available, albeit limited, data suggest that such a decrease is more pronounced in younger patients, whereas in the group of Parkinson’s patients aged 70 years, there is no decrease but an increase in the PPI and PPF. The authors of this study proposed that the excessive modulation of ASR by prepulse in Parkinson’s disease may reflect an impaired integration of sensory inputs, with a lack of coordination between preparatory and executive motor commands[111]. In patients with early Alzheimer’s disease, impaired PPI was observed in 2[112,113] out of 4[112-115] studies, and no patterns of LI length and laterality could be identified. Surprisingly, relatives of Alzheimer’s disease patients showed a PPI increase with a 50 ms LI[113].

DATA OBTAINED UNDER CONDITIONS OF ACTIVATED ATTENTION

Some studies have shown that the effects of selective and non-selective attention on the prepulse modification of SR depend on the length of the LI and may manifest differently in healthy individuals and in psychopathology[3,55-57,59,60,117].

However, the results of these studies are highly dependent on the details of the protocols used, particularly how attention is induced. Until recently, studies using directed attention paradigms have primarily been conducted with healthy participants or patients with schizophrenia spectrum disorders[54-57].

An exception was a study on children with attention deficit hyperactivity disorder. They showed deficits in prepulse modulation in controlled attention conditions[58]. More recently, some researchers have used the PPI under perceived spatial separation (PSS-PPI) paradigm, which relies on the effect of spatial separation of the prepulse and background noise sources to increase attention to the prepulse, compared to the PPI inder perceived spatial co-location (PSC-PPI) paradigm, in which the background noise and prepulse sources are not separated and attention to the prepulse is not increased[35,118-120].

It was found that in the PSC-PPI test, a statistically significant (P = 0.01) PPI deficit was observed only in individuals with first-episode schizophrenia and only at an LI of 60 ms, whereas clinical high-risk participants did not differ from controls. In the PSS-PPI test, a highly significant (P < 0.001) PPI deficit was observed in both patient groups at both 60 ms and 120 ms LIs. In the PSC-PPI test, a statistically significant (P = 0.01) PPI deficit was observed only in first-episode schizophrenia patients and only at 60 ms LIs, whereas clinical high-risk participants did not differ from controls. In the PSS-PPI test, a highly significant (P < 0.001) PPI deficit was observed in both patient groups at both 60 ms and 120 ms LIs[35]. Thus, testing with simultaneous application of PSS-PPI and PSC-PPI allowed us to identify and differentiate PPI deficits in both groups of patients.

Differences were also found only in the PSS-PPI mode, but not in the PSC-PPI mode, when comparing patients with BD and healthy controls for the PPI level at LI 120 ms[118].

However, in another study by the same authors involving three groups of patients: First-episode schizophrenia, BD and major depressive disorder, significant PPI differences from the control level on the PSS-PPI test after Bonferroni correction were observed only in patients with schizophrenia[120]. It can be suggested that the combination of PSS-PPI and PSC-PPI modes may be a promising approach in differential diagnosis.

PREPULSE MODULATION OF NEURAL COMPONENTS OF STARTLE

Some researchers consider the registration of event-related auditory potentials, particularly the N1/P2 complex, as one of the ways to increase the sensitivity and specificity of the assessment of ASR prepulse changes. This approach has been made possible by advances in encephalogram analysis techniques, particularly the reduction of signal contamination[121]. The use of this method significantly increases the sensitivity of PPI estimation, especially when combined with PSS-PPI and PSC-PPI stimulation regimes[119].

In a study on groups of patients with schizophrenia and BD, using a ‘standard’ stimulation protocol and simultaneous EMG and EEG analysis, it was shown that the PPI deficit estimated by EMG parameters was observed only in the schizophrenia group at LI 60 and 120 ms, whereas when assessing the P2-N1 index of auditory evoked potentials, a PPI decrease relative to control was observed in bipolar and schizophrenia patients at LI 30 and 60 ms[122]. Thus, this direction seems promising for finding specific changes in the prestimulus modification of ASR in different disorders and needs to be further developed.

INTEGRATED ASSESSMENT OF PHYSIOLOGICAL PARAMETERS AND GENETIC POLYMORPHISM

Patterns of association between PPI scores and the activity of different brain structures show some degree of specificity in healthy populations and in different psychopathologies[17]. Accordingly, the neurotransmitter mechanisms that determine the specifics of sensorimotor information processing and attention may differ in different psychopathologies. In turn, the activity of neurotransmission in different brain structures depends significantly on the activity of genes encoding receptors and enzymes of neurotransmitter metabolism.

Genetic studies have shown a high degree of heritability of PPI[41], and a meta-analysis of the association between genetic polymorphisms and PPI, performed in a pooled sample of healthy subjects and patients with schizophrenia, found some polymorphic variants with significant effects[123]. One of these polymorphisms, namely rs4680, Val/158Met substitution in the gene catechol-O-methyltransferase (COMT), the main enzyme of dopamine catabolism in the prefrontal cortex, has been investigated in several other studies, also showing its association with PPI scores. Studies in healthy populations showed that the highest PPI levels were observed in methionine homozygotes and the lowest in the Val/Val genotype, while the Val/Met variant showed intermediate levels[123-127]. Polynomial contrast analysis showed a linear relationship between PPI levels and Val allele load[124].

The data on the effect of the rs4680 COMT polymorphism on PPI, obtained in a cohort of patients with schizophrenia, are significantly different from those obtained in healthy individuals.

In a Chinese sample of patients with a first episode of schizophrenia, no effect of rs4680 on PPI was found at LI 120 ms[128]. A study of Russian patients with schizophrenia also found no effect of rs4680 on PPI at LI 60 and 120 ms[125]. In a Caucasian cohort of patients with schizophrenia, Quednow et al[129] found an effect of rs4680 on PPI. In contrast to healthy individuals, genotype-dependent PPI analysis in schizophrenia patients showed a quadratic rather than linear trend, such that higher PPI levels were observed in Met/Met genotype carriers, whereas the difference between Val/Met and Val/Val and between Met/Met and Val/Val was not significant. These results suggest that reduced PPI in Met/Met genotype carriers can be considered a schizophrenia-specific defect, whereas the comparison of PPI levels in Val/Val genotype carriers does not allow us to do so with certainty. Further research in this direction may prove fruitful for the development of a diagnostic tool.

A similar pattern is observed when analyzing the effects of polymorphisms of the gene encoding the alpha-3 subunit of nicotinic receptors, namely polymorphisms rs1051730-cytosine (C) replacement by thymine (T) and rs1317286-adenine (A) replacement by guanine (G). In healthy individuals, the association of rs1051730 and rs1317286 with PPIs was described by a linear trend (CC > CT > TT and AA > GA > GG respectively). At the same time, a significant quadratic trend was observed in patients with schizophrenia with respect to the maximum values of PPIs in CT/GA heterozygotes[130]. These results also suggest that it is important to take genotype into account when comparing PPIs in healthy and schizophrenic patients.

The T102C polymorphism of the serotonin receptor 5-HT2A gene was also found to have opposite associations with PPI levels in healthy and schizophrenic subjects. Patients with the TT genotype have the highest PPI levels, whereas in healthy individuals the presence of the T allele is associated with reduced PPI at LI 50 and, to a lesser extent, at LI 150 ms[131]. Although similar studies have not been performed for other disorders, it could be proposed that polymorphic gene variants that specifically contribute to the regulation of prepulse modification of ASR are also found in these disorders.

CONCLUSION

The phenomenon of prepulse modification of the SR provides an opportunity to study the mechanisms of sensorimotor information processing and attention, as well as their impairment in various neurological and psychiatric disorders. Paradoxically, such a relatively simple behavioral phenomenon has multiple complex regulatory mechanisms. This is related to the importance of SR for the realization of optimal individual behavior in a threatening situation with a high degree of uncertainty. At the same time, the brain structures involved in the regulation of PPI and PPF show different changes in activity in different pathologies, which may lead to changes in the modification of the prepulse that are specific to certain disorders. The detection of such specificity requires the application of uniform schemes, including bilateral recording with separate scoring of data on both sides, the use of comparable sets of LIs, including those that condition SR facilitation, within widely used ‘standard’ methods of stimulation and recording of the SR. It should be noted that elevated PPI levels may be observed in some pathologies. In particular, this has been observed in medication-naive subjects with early psychosis and those at risk of developing psychosis[132], in females with BD[71], in autistic adolescents (Madsen), in elderly patients with Parkinson’s disease[111], and in first-degree relatives of patients with Alzheimer’s disease[113]. Some researchers consider such increases in the PPIs to be evidence of compensatory processes[132]. However, Millian-Morell et al[111], who has observed both PPI and PPF elevations in elderly PD patients, sees these coupled phenomena as evidence of the disintegration of the various stages of sensorimotor information processing. Thus, the inclusion of the long-lead PPF mode appears to be extremely useful in the development of an optimal protocol for ‘standard’ testing of the prepulse modification of the SR. EEG assessment, including evoked potentials and changes in oscillation patterns, appears to be a promising new approach to estimating PPI and PPF. The application of new modes of stimulation, including the paradigm of pattern separation, could also be very useful. Finally, an integrated assessment of physiological parameters and genetic polymorphisms is proposed as very useful both for diagnosis and for delineating the mechanisms of neuropsychiatric disorders. The evidence to date suggests that the use of the SR and its prepulse modification in a battery of neurophysiological screening tools can increase the efficiency of identifying people in need of psychological and psychiatric care, assessing the dynamics of the state and the impact of therapy. The achievement of an acceptable level of sensitivity and specificity of such neurophysiological test systems can be ensured by a combination of different paradigms providing, in addition to the assessment of sensorimotor gating and integration, the assessment of sensory filtering (suppression of the evoked potential P50), as well as various indicators of voluntary attention (evoked potentials P300 and N400) and executive control (antisaccade test, continuous performance test, etc.). Furthermore, the analysis of data arrays accumulated over many years in studies of different cohorts of subjects, as well as the results of recent studies using new experimental approaches, indicate the possibility of increasing the sensitivity and specificity of the method of prepulse modification of the SR for the diagnosis, prognosis and testing the effectiveness of treatment of mental and neurological disorders.