Fear conditioning is preserved in very preterm-born young adults despite increased anxiety levels

doi:10.21203/rs.3.rs-2624330/v1

Download PDF

Article

Fear conditioning is preserved in very preterm-born young adults despite increased anxiety levels

https://doi.org/10.21203/rs.3.rs-2624330/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 12 Jul, 2023

Read the published version in Scientific Reports →

You are reading this latest preprint version

Very preterm birth is associated with an increased risk for anxiety disorders. Abnormal brain development may result in disordered fear learning processes, which may be exacerbated by environmental risk factors and persist in adulthood. We tested the hypotheses that very preterm-born young adults displayed higher levels of fear conditioning, less differentiation between threat, CS+, and safety, CS-, signals and stronger resistance to extinction relative to term-born controls. A group of 37 very preterm-born young adults and 31 age- and sex-matched term-born controls performed a differential fear conditioning paradigm on two consecutive days. Acquisition and extinction training were performed on day 1. Recall and reinstatement were tested on day 2. Preterm-born participants showed significantly higher levels of anxiety in the Depression-Anxiety-Stress-Scale-21 questionnaire. The fear conditioning outcome measures, skin conductance response amplitudes and anxiety ratings, were overall higher in the preterm-born group compared to controls. Acquisition, extinction, recall and reinstatement of differential conditioned fear responses, CS+ > CS-, however, were not significantly different between the groups. There were no significant group by stimulus type interactions. The finding of preserved associative fear learning in very preterm-born young adults was unexpected and needs to be confirmed in future studies.

Biological sciences/Neuroscience

Biological sciences/Neuroscience/Learning and memory/Extinction

Biological sciences/Neuroscience/Learning and memory/Fear conditioning

Prematurity

anxiety

fear

classical conditioning

associative learning

Substantial progress has been made in obstetrical and neonatal care of very preterm-born infants over the last decades (Ancel et al., 2015). As a consequence, the rate of mortality and and severe morbidity leading to cerebral palsy has been significantly lowered (de Vries et al., 2011; Platt et al., 2007). However, cognitive impairments as well as emotional and behavioral problems have become more apparent, and are reported in up to 50% of very preterm-born children and adolescents (Lammertink et al., 2021). Following the vulnerability hypothesis, mental disorders with early manifestation in childhood and adolescence are a frequent progenitor of mental disorders in adulthood (Steinhausen, 2013). Prematurity is a risk factor for emotional and behavioral disorders throughout life, and the relative frequency seems to be increasing (Lammertink et al., 2021).

More specifically, very preterm-born children and adolescencents have a significant higher risk for attention-deficit disorder with and without hyperactivity (ADD/ADHD) and autism spectrum disorders (ASD) compared to their term-born peers (Agrawal et al., 2018; Breeman et al., 2016; D'Onofrio et al., 2013; Johnson and Marlow, 2011). Furthermore, very preterm birth (≤ 32 weeks of gestation) is associated with a 2–4 times increased risk for emotional problems such as anxiety disorders, depression, and obsessive-compulsive disorders (Frazier et al., 2021; Gilchrist et al., 2021; Johnson and Marlow, 2011; Lærum et al., 2017; Gustafsson et al., 2009; Van Lieshout et al.,2015). The co-occurrence of symptoms associated with ADD/ADHD, ASD and anxiety disorders has been denoted as "preterm behavioral phenotype", and occurs mainly in children with very low birth weight (Bindt, 2022).

Aberrant brain development likely plays a role associated with diffuse white matter disease, a thin corpus callosum, enlarged lateral ventricles and a reduction of cerebellar volume (Volpe, 2009a). This aberrant brain development has been summarized under the term “encephalopathy of prematurity” by Volpe (Volpe, 2009a). Environmental factors may also contribute to the development of emotional and behavioral problems. Very preterm-born infants experience a stressful peri- and postnatal period with repeated painful procedures, sleep interruptions, an artificial environment and a long-lasting separation from their parents (Anderson et al., 2015; Ortinau et al., 2015; Volpe, 2009b).

Abnormal learning processes are thought to play an important role in the development of mental disease (Cohen et al., 2022). For example, anxiety disorders are frequently related to disordered acquisition and extinction of conditioned fear responses (Lissek et al., 2005). It is conceivable that the abnormal brain development results in disordered emotional learning processes, which are exacerbated by environmental risk factors and persist into aduldhood. Initial studies in rodent models of prematurity have reported altered fear conditioning (Matsuda et al., 2021; Pierre et al., 2022; Xia et al., 2020). Furthemore, a reduced volume of the amygdala has been shown in very preterm-born infants, an important hub of the fear conditioning circuitry (Chau et al., 2019; Schmitz-Koep et al., 2021). As yet, it is unknown whether young adults born preterm have deficits in the acquisition and extinction of learned fear responses. Based on previous findings in patients with anxiety disorders, we tested the hypotheses that very preterm-born young adults display higher levels of fear conditioning (Duits et al., 2017), less differentiation between the CS+ (that is, threat) and the CS- (that is, safety) signals (Britton et al., 2011) and stronger resistance to extinction relative to healthy controls (Duits et al., 2021).

Depression-Anxiety-Stress-Scale-21 (DASS-21) questionnaire

Scores differed between preterm and control groups especially in the anxiety scale (Fig. 2). Sixteen (43.2%) preterm participants showed mildly to moderately elevated scores of anxiety, but only four (12.9%) control participants. Nine of the twelve preterm participants, who had a history of emotional disorders, were part of the subgroup with elevated anxiety scores. Six (16.2%) preterm participants and one (3.2%) control presented with mildly elevated depression scores. Ten (27%) preterms and one (3.2%) control showed mildly elevated stress scores. Anxiety (U = 399.5, p = 0.032, Mann-Whitney-U test) and depression scores (U = 388.5, p = 0.023) were significantly higher in the preterm group compared to controls. Stress scores were not significantly different between groups (U = 490, p = 0.31).

Skin conductance responses

Group mean SCR amplitudes in the habituation phase, early and late acquisition and extinction phases on day 1, and early and late recall and reinstatement phases on day 2 are shown in Fig. 3. Statistical findings are summarized in Table 2. Throughout the experiment, mean SCR amplitudes were numerically higher in the preterm group (red columns) compared to the control group (blue columns). The group difference was significant in reinstatement phase. No other significant difference between the preterm and control groups was observed.

Habituation phase (day 1)

Mean SCR amplitude was higher towards the CS + E compared to the CS + U and CS- in both groups (Fig. 3; Table 2). The reason for this difference was that the CS + E was always shown first, and both preterms and controls showed a significantly higher SCR amplitude to this very first trial compared to all other habituation trials (see Fig. S1, Table S1 in Supplementary Materials). Importantly, at the end of the habituation phase there was no significant difference in SCR amplitude between the three stimuli (CS + E, CS + U and CS-). SCRs in the first CS + E trial was significantly higher than CS⁺U and CS⁻ (all p values < 0.001, least squares means test) in both groups, but not in the second trial (all p values > 0.999, least squares means test).

Acquisition phase (day 1): Both groups learned to differentiate the CS + U and the CS + E from the CS- in the acquisition phase with significantly higher SCR amplitudes towards the two CS + s compared to the CS- (Fig. 3, Table 2; see also Fig. S1 and Table S1 for individual trial data). This difference was more prominent during late acquisition in both groups. As expected, SCR amplitudes declined in late compared to early acquisition, with SCR habituation being a common finding in fear conditioning studies (Batsikadze et al., 2022; Inoue et al., 2020; Merz et al., 2016). Non-parametric ANOVA-type statistics revealed significant main effects of stimulus (CS⁺E vs CS⁺U vs CS⁻: F_(1.77,108) = 16.66, p < 0.001) and block (early vs late phase; F_(1,58.1) = 90.37, p < 0.001), but not of group (p = 0.482). None of the interactions were significant: block × stimulus (p = 0.886), group × stimulus (p = 0.886), group × block (p = 0.090), group × stimulus × block (p = 0.504).

Extinction phase (day 1): At the beginning of the extinction phase, SCR amplitudes were higher towards the CS + E compared to the CS- in both groups, confirming that participants had learned to differentiate between the CS + E and CS- in the acquisition phase. Both groups learned to extinguish the conditioned responses, with no significant difference between the CS + E and CS- in late extinction. Non-parametric ANOVA-type statistics revealed significant main effects of block (early vs late phase; F_(1,62.6) = 29.79, p < 0.001), stimulus type (CS + E vs CS-; F_(1,61.5) = 12.35, p = 0.001) and a significant block x stimulus (F_(1,55.7) = 7.59, p = 0.008) interaction. The group main effect (p = 0.513) and all other interactions were not significant: group × stimulus (p = 0.730), group × block (p = 0.210) or group × stimulus × block (p = 0.236).

Recall phase (day 2)

In early recall, SCR amplitudes were higher in CS + E and CS + U blocks compared with CS- blocks in both groups. In late recall, SCR amplitudes did not differ between CS + s and CS- in both groups. Non-parametric ANOVA-type statistics revealed significant main effects of block (early vs late phase; F_(1,64.6) = 56.42, p < 0.001) and stimulus type (CS + E vs CS + U vs CS−; F_(1.87,116) = 6.93, p = 0.002) and a significant block x stimulus interaction (F_(1.81,106) = 8.82, p < 0.001). The group main effect (p = 0.099) and all other interaction effects were not significant [group × stimulus (p = 0.223), group × block (p = 0.922) or group × stimulus × block (p = 0.235)]. In early recall, SCRs in CS + E and CS + U blocks were significantly different from CS- block in both groups (all p values < 0.013, least squares means test). No difference between CS + E and CS + U were revealed (p = 0.870, least squares means test).

Reinstatement phase (day 2)

SCR values were higher in the preterm group compared with controls towards all CSs. In early reinstatement, SCR amplitudes were numerically higher towards the CS + s compared to the CS- in both groups. Non-parametric ANOVA-type statistics revealed a significant main effect of group (preterm vs control; F_(1,65.9) = 6.96, p = 0.01). Main effects of block (early vs late phase; F_(1,63) = 38.28, p < 0.001) and block x stimulus interaction (F_(1.88,104) = 3.89, p = 0.023) were significant. No significant main effect of stimulus (p = 0.148), and no significant interaction effects [group × stimulus (p = 0.660), group × block (p = 0.414), group × stimulus × block (p = 0.287)] were revealed. SCRs in preterm group were significantly higher compared to SCRs in the control group throughout the whole reinstatement phase (p = 0.01, least squares means test).

Table 1

Group characteristics of very preterm adults and controls
	Very preterm (n = 37)	Term (n = 31)	p^a
Clinical characteristics
Gestational age, weeks [range]	29 + 2.8 [26 + 2–32.0]	39 + 5.7 [37.0–42.0]	< .001
Birth weight, grams [range]	1264.2 [520–2370]	3637.5 [2995–5360]	< .001
Female, n (%)	19 (51.4%)	16 (51.6%)	0.983
IVH < grad III, n (%)	1 (2.7%)	0 (0%)	1
Follow-up characteristics
Age at assessment, years [range]	20.2 [17.8–27]	22.4 [18–29]	< .001
Education, high^b, n	30	29	0.131
Parental education, high^b, n	22	24	0.115
Health status^c	5.88 [4.14–7]	5.47 [3.71–6.57]	0.015
IQ	97 [ 83–123]	108.71 [89–122]	< .001
Any therapy^d, n (%)	15 (40.5%)	0 (0%)	< .001
Any psychiatric/ social-emotional disorders^e, n (%) - Phobias/ anxiety/ depression, n (%) - ADS/ ADHS, n (%) - Difficulties social-emotional adaptation, n (%)	12 (32.4%) 5 (13.5%) 5 (13.5%) 2 (5.4%)	1 (3.2%) 0 (0%) 0 (0%) 0 (0%)	0.002
DASS-21 depression, n (%)	6 (16.2%)	1 (3.2%)	0.023
DASS-21 anxiety, n (%)	16 (43.2%)	4 (12.9%)	0.032
DASS-21 stress, n (%)	10 (27%)	1 (3.2%)	0.31
Developmental disorders^f, n (%)	15 (40.5%)	0 (0%)	< .001
Notes. Data are presented as mean (standard deviation) if not indicated otherwise. ^at-test or Mann-Whitney U-test and chi-square results or Fischer’s exact test for continuous and categorical data, respectively. ^b > 10 years school. ^c assessment of health status based on the Life Satisfaction Questionnaire (Fragebogen zur Lebenszufriedenheit, FLZ) ^d having any therapies, including speech therapy, physical therapy, or occupational therapy. ^e psychiatric disorders including attention-deficit-(hyperactivity)-disorder, emotional disorder. SGA - Small for gestational age (birth weight < 10th percentile), IVH - Intraventricular haemorrhage, BPD - Bronchopulmonary dysplasia, ROP - Retinopathy of Prematurity, IQ - Intelligence quotient based on the Wechsler Adult Intelligence Scale – Third Edition ^f developmental disorders concerning language, gross- or fine motor functions which needed therapy in the past (speech therapy, physical therapy or occupational therapy)

Table 2

Results of the non-parametric ANOVA-type statistics for repeated measures for skin conductance response (SCR) amplitudes and valence, arousal, fear and US expectancy ratings comparing the groups of all preterm participants and controls.
Factor	Numerator Df	Denominator Df
Skin conductance responses
Habituation
Stimulus Group Stimulus*Group	1.88 1 1.88	103 65.9 103	13.86 1.75 1.53	< .001*** 0.191 0.221
Fear acquisition training
Stimulus Block Group StimulusBlock BlockGroup StimulusGroup StimulusBlock*Group	1.77 1 1 1.94 1 1.77 1.94	108 58.1 65.5 116 58.1 108 116	16.66 90.37 0.50 0.11 2.98 0.10 0.68	< .001* < .001* 0.482 0.886 0.090 0.886 0.504
Extinction training
Stimulus Block Group StimulusBlock BlockGroup StimulusGroup StimulusBlock*Group	1 1 1 1 1 1 1	61.5 62.6 66 55.7 62.6 61.5 55.7	12.35 29.79 0.43 7.59 1.61 0.12 1.44	< .001* < .001* 0.513 0.008* 0.210 0.730 0.236
Recall
Stimulus Block Group StimulusBlock BlockGroup StimulusGroup StimulusBlock*Group	1.87 1 1 1.81 1 1.87 1.81	116 64.6 65.7 106 64.6 116 106	6.93 56.42 2.81 8.82 0.01 1.52 1.47	0.002* < .001* 0.099 < .001* 0.922 0.223 0.235
Reinstatement
Stimulus Block Group StimulusBlock BlockGroup StimulusGroup StimulusBlock*Group	2 1 1 1.88 1 2 1.88	118 63 65.9 104 63 118 104	1.94 38.28 6.96 3.89 0.67 0.42 1.26	0.148 < .001*** 0.010* 0.023* 0.414 0.660 0.287
Valence
Group Stimulus Time GroupStimulus GroupTime StimulusTime GroupStimulus*Time	1 1.55 3.42 1.55 3.42 5.12 5.12	64.8 97.9 210 97.9 210 306 306	0.08 66.44 24.88 0.94 0.86 37.94 1.00	0.785 < .001* < .001* 0.373 0.473 < .001*** 0.420
Arousal
Group Stimulus Time GroupStimulus GroupTime StimulusTime GroupStimulus*Time	1 1.51 3.37 1.51 3.37 4.75 4.75	64.5 90.7 215 90.7 215 285 285	2.71 75.97 39.94 0.46 1.07 30.06 0.43	0.104 < .001* < .001* 0.578 0.367 < .001*** 0.815
Fear
Group Stimulus Time GroupStimulus GroupTime StimulusTime GroupStimulus*Time	1 1.42 3.22 1.42 3.22 4.78 4.78	64.5 82 203 82 203 274 274	4.00 48.44 44.69 1.16 0.73 26.88 0.71	0.050* < .001* < .001* 0.304 0.547 < .001*** 0.608
US expectancy
Group Stimulus Time GroupStimulus GroupTime StimulusTime GroupStimulus*Time	1 1.69 3.31 1.69 3.31 5.01 5.01	65 95.2 207 95.2 207 301 301	1.76 87.34 70.50 0.87 1.35 41.07 0.66	0.189 < .001* < .001* 0.405 0.257 < .001*** 0.658

* Significant results at p < 0.05.

*** Significant results at p < 0.001.

Questionnaires

Valence, arousal, fear. Prior acquisition training there was no significant difference in ratings of the CS + s and the CS- in both groups. Post acquisition training valence of the CS + s was rated significantly less pleasant compared to the CS-, arousal and fear were rated significantly higher (Fig. 4A-C, all p values > 0.001, see Table 2 for summary of non-parametric ANOVA-type statistical analysis). Differences between CS + s and CS- ratings remained post extinction, post recall and post reinstatement. Post hoc tests showed significantly less pleasant valence rating, significantly higher arousal and fear ratings towards CS + E and CS + U compared to CS- post acquisition, post extinction training, post recall and post reinstatement (least squares means tests, all p values < 0.001), but not prior acquisition training (least squares means test, all p values > 0.06). Main effects of time (prior acquisition vs post acquisition vs post extinction vs post recal vs post reinstatement) were significant (all p values < 0.001). Regarding fear ratings the main effect of group was significant (controls vs preterms: F_(1,64.5) = 4.00, p = 0.05). Pre-acquisition fear ratings were significantly higher in preterms than controls (p = 0.027), but not valence and arousal ratings (all p values > 0.382). Post hoc tests showed no significant group main effects and stimulus x group interactions post acquisition, extinction, recall and reinstatement (all p values > 0.08; non-parametric ANOVA on individual time points).

US unpleasantness, US expectancy, CS-US contingency. Preterm and control participants rated the US as unpleasant and unpleasantness was not different between groups [post acquisition training: preterms: median 7 (IQR 7–8); controls: median 8 (IQR 7–8); on a Likert scale from 1 (not unpleasant) to 9 (very unpleasant); Mann-Whitney U test: U = 482.5, p = 0.347].

US expectancy, as an index of conscious CS-US contingency awareness (Lonsdorf et al., 2017), was not different between groups. Both groups learned the CS-US contingencies.

Prior acquisition training, US expectancy ratings after CS + E, CS + U and CS- presentations were not different. Post acquisition US expectancy after the CS + E and CS + U were rated significantly higher compared to the CS- (all p values < 0.001). No significant differences between groups were revealed post acquisition, extinction and recall (all p values > 0.256; non-parametric ANOVA on individual time points). Post-reinstatement US expectancy ratings were significantly higher in preterms than in controls (p = 0.035; Fig. 4D, Table 2).

Post acquisition training, one control (3.2%) and nine preterms (24.3%) reported that they did not recognize a pattern between CS + and US presentations. The rest of participants reported that they recognized a pattern between CS + and US contingency after 3.76 ± 2.36 mins. There were no significant differences between groups (Mann-Whitney U test; U = 327, p = 0.150). To analyze in more detail whether learning of the CS-US contingencies differed between control and preterms, who recognized a pattern, a compound contingency score was calculated considering the following four questions, which were asked at the end of the acquisition phase: i) estimate the percentage of US after CS + E; ii) estimate the percentage of US after CS + U; iii) estimate the percentage of US after CS; iv) which color (CS) was never followed by an US. The subscores for the first three questions were calculated by how far they were from the correct answer in 10% steps and were rated between 1 (most accurate) and 0 (most inaccurate). The question 4 had a binary 0 or 1 score, as there were only correct and incorrect answers. The overall compound score is the sum of the above subscores and ranged from 0 (wrong answer to all questions) to 4 (correct answer to all questions). Preterm born participants scored significantly below controls (Fig. 5; Mann-Whitney U test; U = 351.5, p = 0.01).

Comparing anxious and non-anxious preterm participants

In the following we tested whether preterm participants who scored in the DASS21 anxiety subscore in the abnormal range performed differently from preterms who scored within the normal range (16 anxious and 21 non-anxious preterm participants based on a cut off > 7; see Fig. 2). The control group consisted of the 27 participants who scored within normal range (non-anxious control group), with four controls performing above.

SCR amplitudes. Results remained largely the same and showed only minor deviations from the comparison of all preterm participants and non-anxious controls. Again, mean SCR amplitudes were numerically higher in preterms compared to non-anxious controls. Mean SCR amplitudes, however, were not different between the anxious and non-anxious preterm subgroups. Main group effects were not significant. The group difference in the reinstatement phase was close to significant (p = 0.088). No other significant differences between the preterm and non-anxious control groups were observed. Group mean SCR amplitudes in anxious preterms, non-anxious preterms and non-anxious controls in early and late blocks in the habituation, acquisition, and extinction phases on day 1 and the recall and reinstatement phases on day 2 are shown in Fig. 6(see also Figure S2 for individual trial data). Statistical findings are summarized in Table 3.

Questionnaires. Data is shown in Fig. 7 and statistical findings are summarized in Table 3. Regarding fear ratings the main effect of group was significant (non-anxious controls vs non-anxious preterms vs anxious preterms: F_(1.87,48.9) = 3.90, p = 0.029). There was no significant stimulus x group interaction (p = 0.253). Main effect of time (prior acquisition vs post acquisition vs post extinction vs post recall vs post reinstatement) was significant (p < 0.001). Post hoc non-parametric ANOVA-type statistics revealed numerically higher fear ratings in anxious preterms compared to non-anxious controls, but group effect was not significant (prior acquisition: p = 0.1; all other p values > 0.266).

Post acquisition CS + s were rated more unpleasant than CS-. The difference remained in subsequent phases. Main effect of time (prior acquisition vs post acquisition vs post extinction vs post recall vs post reinstatement) was significant (p < 0.001). Post hoc non-parametric ANOVA-type statistics revealed a close-to-significant effect of group in post extinction arousal (p = 0.057), and a significant group x stimulus interaction in post-recall valence (p = 0.042). Least squares means tests revealed significantly more negative post-recall valence ratings in anxious preterms and non-anxious controls towards CS + s compared to the CS- (all p values < 0.002), but not in non-anxious preterms (all p values > 0.98).

US expectancy ratings were not significantly different between groups considering all experimental phases together. Post hoc non-parametric ANOVA-type statistics revealed a significant post-recall group x stimulus interaction (p = 0.027). Least squares means tests revealed significantly higher US expectancy ratings in anxious preterms and non-anxious controls towards CS + s compared to the CS- (all p values < 0.001), but not in non-anxious preterms (all p values > 0.08).

Post acquisition training, 1 non-anxious control (3.7%), 3 anxious (18.8%) and 6 non-anxious (26.6%) preterms reported that they did not recognize a pattern between CS + and US presentations. Regarding the CS-US contingency compound score in the participants, who recognized a pattern, post hoc testing showed that anxious preterms did not significantly differ from non-anxious preterms (U = 154, p = 0.857). Non-anxious preterms performed significantly below non-anxious controls (U = 177, p = 0.047), and the difference was close to significance comparing anxious preterms and non-anxious controls (U = 141.5, p = 0.063) (Fig. 5B).

Table 3

Results of the non-parametric ANOVA-type statistics for repeated measures for skin conductance response (SCR) and valence, arousal, fear and US expectancy ratings comparing the groups of anxious preterms, non-anxious preterm and non-anxious controls.
Factor	Numerator Df	Denominator Df	F	P
Skin conductance responses
Habituation
Stimulus Group Stimulus*Group	1.94 1.91 3.72	107 47.5 107	12.74 1.34 1.33	< .001*** 0.271 0.265
Fear acquisition training
Stimulus Trial Group StimulusTrial TrialGroup StimulusGroup StimulusTrial*Group	1.82 1 1.94 2 2 3.57 3.44	94.4 59.1 50.4 73.8 59.1 94.4 73.8	15.51 81.62 0.27 0.04 1.47 0.41 0.65	< .001* < .001* 0.760 0.963 0.237 0.780 0.609
Extinction training
Stimulus Trial Group StimulusTrial TrialGroup StimulusGroup StimulusTrial*Group	1 1 1.92 1 1.84 1.91 1.89	48.1 42.5 48.1 46.4 42.5 48.1 46.4	13.89 23.72 0.23 10.37 0.92 2.03 1.47	< .001* < .001* 0.785 0.002* 0.398 0.145 0.241
Recall
Stimulus Trial Group StimulusTrial TrialGroup StimulusGroup StimulusTrial*Group	1.84 1 1.93 1.73 1.94 3.53 3.13	90.3 49.1 48.1 72.2 49.1 90.3 72.2	5.62 52.10 1.63 4.91 0.48 0.93 0.76	0.006* < .001*** 0.208 0.013* 0.614 0.444 0.524
Reinstatement
Stimulus Trial Group StimulusTrial TrialGroup StimulusGroup StimulusTrial*Group	1.96 1 1.93 1.92 1.97 3.47 3.6	79.9 52.6 48.7 99.3 52.6 79.9 99.3	1.28 40.32 2.57 3.83 0.20 0.36 2.19	0.284 < .001*** 0.088 0.027* 0.815 0.808 0.082
Valence
Group Stimulus Time GroupStimulus GroupTime StimulusTime GroupStimulus*Time	1.9 1.66 3.41 3.2 6.19 5.29 9.08	47.3 83.4 155 83.4 155 206 206	0.03 63.62 23.36 1.76 0.91 32.15 1.19	0.970 < .001* < .001* 0.153 0.495 < .001*** 0.301
Arousal
Group Stimulus Time GroupStimulus GroupTime StimulusTime GroupStimulus*Time	1.93 1.58 3.2 3.04 5.76 5.14 8.58	53.3 78.4 135 78.4 135 189 189	2.84 74.47 33.07 0.68 0.82 24.64 1.42	0.069 < .001* < .001* 0.566 0.550 < .001*** 0.187
Fear
Group Stimulus Time GroupStimulus GroupTime StimulusTime GroupStimulus*Time	1.87 1.52 3.09 2.9 5.53 5.12 9.08	48.9 74.4 132 74.4 132 227 227	3.90 47.46 38.68 1.39 0.44 26.07 1.48	0.029* < .001* < .001* 0.253 0.841 < .001*** 0.157
US expectancy
Group Stimulus Time GroupStimulus GroupTime StimulusTime GroupStimulus*Time	1.92 1.89 3.32 3.47 5.98 4.85 8.13	49.7 84 141 84 141 173 173	2.23 78.53 58.09 0.40 0.76 34.96 1.91	0.120 < .001* < .001* 0.784 0.606 < .001*** 0.056

* Significant results at p < 0.05.

*** Significant results at p < 0.001.

The aim of the present study was to evaluate fear conditioning in very preterm born young adults, a patient group at risk for the development of anxiety disorders. Acquisition and extinction learning was largely preserved, except for slightly limited learning of the CS-US contingencies. Recall and reinstatement of learned fear responses after successful extinction training were not significantly different from term-born young adults. Anxiety ratings and skin conductance amplitudes were generally higher in very preterm-born young adults, but these findings were unspecific and not linked to aberrant fear-related associations.

Preterms show high levels of anxiety.

A higher percentage of the very preterm-born young adults than controls showed elevated depression, anxiety and stress scores in the Depression-Anxiety-Stress-Scale-21 (DASS-21). Our observations that very preterm-born young adults are more anxious match well with the literature (Eryigit-Madzwamuse et al., 2015; Johnson and Marlow, 2011; Johnson et al., 2019; Pesonen et al., 2008). Preterm birth is associated with a higher risk for attention-deficit/hyperactivity disorder, mood and anxiety disorders and psychiatric disorders (Breeman et al., 2016; Lærum et al., 2017; Van Lieshout et al., 2015). Preterm-born adults show significantly more symptoms of inattention and anxiety, positively correlating with lower gestational age. As a consequence Johnson and Marlow introduced the concept of a “preterm behavioral phenotype” characterized by a higher risk for symptoms of inattention, anxiety and social difficulties (Johnson and Marlow, 2011). Preterm-born adults score higher in avoidant personality scales (Johnson et al., 2019) and are less extraverted and more withdrawn (Eryigit-Madzwamuse et al., 2015; Pesonen et al., 2008).

Increased anxiety levels in the DASS21 make it less likely that the largely negative findings of the present study are explained by a lack of anxious phenotype in our study population. Furthermore, results did not change when only the very preterm-born subgroup who rated within the abnormal range in the anxiety subscore of the DASS21 was investigated.

Acquisition and extinction of conditioned fear responses is preserved in preterms.

There is evidence in the literature that the level of fear responses towards threat stimuli (that is, the CS+) is increased in anxiety disorders (Duits et al., 2017), and that the differentiation between the threat and safety (that is, the CS-) stimuli is less (Britton et al., 2011). Lack of differentiation is of interest, because one potential underlying mechanism of anxiety disorders might be an elevated fear response to safety cues in the context of threat cues (Craske et al., 2012). In fear conditioning paradigms individuals with anxiety disorders have been found to show higher skin conductance responses to the CS- compared to controls (Craske et al., 2008; Lissek et al., 2005; Lissek et al., 2009; Waters et al., 2009). This has been interpreted as an impaired learning of inhibition of fear to safety cues (Gewirtz et al., 1997), which might be partly caused by a greater stimulus generalization between CS + and CS- cues (Lissek et al., 2010).

We therefore tested the hypotheses that very preterm-born young adults display higher levels of learned fear responses to threat cues, and less differentiation between the threat and safety cues. In both cases, a significant interaction between conditioned stimulus type (CS+, CS-) and group is to be expected. This, however, was not the case. Both the very preterm-born and the term-born groups showed significantly higher fear responses towards both CS + s as compared to the CS- in the early and late acquisition phases. This difference remained in early extinction phase and vanished during late extinction. However, there were no significant stimulus type by group, or stimulus type by group by block interaction effects, neither considering SCR amplitude nor ratings of fear, arousal and valence. SCR amplitudes appeared generally increased in preterms including the initial habituation phase, but the group effects did not become significant. Likewise, fear ratings were generally higher in preterms compared to controls. Comparing all preterms and all controls, the group difference was significant at the time prior acquisition. The main effect of group was also significant comparing anxious preterms, non-anxious preterms and non-anxious controls (independent of the phase). Thus, there was an unspecific increase of anxiety ratings.

The lack of significant abnormalities in the acquisition and extinction of differential conditioned fear responses in very preterm-born young adults was unexpected. Negative findings, however, are in accordance with metanalyses of fear conditioning studies in patients with different anxiety disorders showing that initial findings are frequently difficult to replicate and abnormalities are frequently weak (Duits et al., 2015; Lissek et al., 2005).

Findings of preserved acquisition of conditioned fear responses are also at variance with a previous study of our group showing a significantly reduced ability to acquire conditioned eyeblink responses in very preterm-born children and young adults (Tran et al., 2017), another form of aversive associative learning. Delay eyeblink conditioning is strongly cerebellar dependent and prematurity has a significant detrimental effect on the cerebellum (De Zeeuw et al., 2005; Gerwig et al., 2003; McCormick et al., 1984). Important developmental processes of the cerebellum are interrupted by prematurity leading to cerebellar growth failure (Anderson et al., 2015; Ortinau et al., 2015; Messerschmidt et al., 2005; Tam, 2013; Volpe, 2009b,c). The cerebellum, on the other hand, is not only involved in eyeblink conditioning, but also in fear conditioning (Batsikadze et al., 2022; Ernst et al., 2019; Sacchetti et al., 2005). However, fear conditioning relies on a more widespread fear learning circuitry including the amygdala (Kim et al., 2011; Linnman et al., 2012; Milad et al., 2007a,b). Amygdala and related fear circuitry are likely to be also affected by prematurity (Chau et al., 2019; Ortinau et al., 2015; Schmitz-Koep et al., 2021), but possibly to a lesser extent as the cerebellum (Volpe, 2009a-c). Likewise, fear conditioning is only mildly affected in cerebellar disease, and significantly less compared to the detrimental effects on eyeblink conditioning in humans (Maschke et al., 2002).

Awareness of CS-US contingencies is reduced in preterms.

Mean US expectancy ratings towards the CS + s und CS- were not significantly different comparing preterms and control participants. The percentage of preterms, however, was less who reported that they recognized a pattern between the CS + and US presentation after the acquisition phase. Furthermore, in preterms who recognized a pattern, contingency ratings were significantly less accurate compared to controls. Thus, knowledge of the CS-US contingencies was less in very preterm-born young adults at the end of acquisition training. Contingency awareness is likely linked to attention and working memory processes (Dawson et al., 1976). Very preterm-born participants have a high risk of attention and learning problems (Breeman et al., 2016; Johnson and Marlow, 2011) which has been associated with cerebellar growth failure (Murray et al., 2014; Omizzolo et al., 2014). Awareness of the CS-US contingencies is considered a prerequisite for conditioned skin conductance responses to occur (Lonsdorf et al., 2017). The observed deficits in awareness, however, were small. The known variability of SCRs may have prevented the detection of mild fear conditioning deficits in preterms. The observation of reduced CS-US contingency awareness warrants future studies in larger preterm populations.

Unaltered recall and reinstatement of learned fear following extinction in preterms.

Extinction learning does involve more than erasure. There is new inhibitory learning going on, that is participants learn that the CS + is no more followed by an aversive stimulus (Lattal et al., 2006). A stronger resistance to extinction relative to healthy controls is thought to contribute to the development of anxiety disorders (Duits et al., 2021). Extinction learning, however, was not significantly different between preterms and controls.

Phenomena like spontaneous recovery, renewal and reinstatement provide evidence than the initial fear memory is not erased during extinction learning, and are possible reasons why exposure therapy shows frequently return of fear (Pavlov, 2010; Robbins, 1990). During tests of recall and reinstatement, there is competition between the learned extinction memory and the learned fear memory (Bouton et al., 1979; Bouton et al., 1983). Return of learned fear following extinction learning has been shown to be stronger in patients with anxiety disorders compared to controls (Bouton et al., 1979; Milad et al., 2007a; Rescorla et al., 1975).

In early recall, both preterms and controls showed return of differential fear responses, with higher fear responses to the CS + compared to the CS-, which vanished in late recall. Neither controls nor preterms showed a significant difference between the CS+, which had been extinguished and the CS + which had not been extinguished. Thus, both groups showed high levels of spontaneous recovery of learned fear towards the extinguished CS+. Similar to acquisition and extinction, SCR amplitudes were numerically higher in preterms. There were, however, no significant stimulus type (CS + E vs CS + U vs CS-) by group interactions. Comparable results were obtained in testing of reinstatement effects, that is reinstatement of fear responses after (unpaired) US presentation following the renewed extinction of fear responses. There were no significant differences in SCR responses towards the CS + s and the CS-, that is, we saw generalized reinstatement effects (in both groups) which is frequently observed in humans (Haaker et al., 2014) Furthermore, the general increase of SCR responses in preterms compared to controls observed throughout the experiment became significant. In addition, post-reinstatement US expectancy ratings were significantly higher in preterms. However, considering the early and late reinstatement, but also comparing late recall phase and early reinstatement phase, there were no significant phase by group interactions (Table S2 in Supplementary Materials). Thus, return of fear responses in recall and reinstatement were not significantly different between groups. We were therefore unable to provide evidence that return of fear following extinction learning is elevated in prematurity.

Limitations.

Lack of differences between very preterm-born and term-born participants may be explained by several reasons. First, the fear conditioning paradigm used in the present study was complex. It involved the presentation of two contexts, one presented during acquisition training, the other during extinction training, recall and reinstatement. This was done to control for context-related associations, which is commonly done in the rodent literature, but rarely in the human literature (Haaker et al., 2014; Lonsdorf et al., 2017; Milad et al., 2007a,b). The contexts were pictures of two different office spaces shown on a monitor to the participants. This is likely different from testing in two different chambers, which is usually done in rodent studies. The strength of the learned associations between the CS + s and the US, and the CS- and safety may have been less, because participants learned in addition the association with a complex context. Likewise, the use of two CS + s, one which got extinguished and the other not, may have hampered the strength of the CS+/US associations. The use of different contexts and two CS + s may have increased variability of outcome measures and impeded the detection of small differences between preterms and controls. Future studies are needed using less complex fear conditioning paradigms, for example using a single CS + and a single CS-, which are presented in the same neutral context, to confirm the present results.

Furthermore, the use of a complex paradigm might have caused a partial suppression of the autonomic nervous system/ emotional behavior of the participants. When cognitive tasks are new and challenging, a suppression of autonomic nervous system takes place for frontal circuits to allocate neural resources according to cognitive demand (Chen et al., 2020; Lövdén et al., 2010). Again, this may have hampered detection of small differences in SCR between groups.

In addition, fear conditioning abnormalities may be more prominent in preterm participants presenting with pronounced anxiety or other emotional disorders. However, as outlined above, twelve very preterm-born young adults presented with a history of emotional disorders during adolescence, and a higher percentage of very preterm-born participants (43.2%) showed mild to moderate levels of anxiety compared to controls (12.9%) based on the DASS-21.

Finally, abnormalties may have been missed because of reduced autonomic responses in preterm born individuals. SCR reflects the activity of the sympathetic axis of autonomic nervous system since eccrine sweat glands are innervated by efferent fibers of the sympathetic nerves (Critchley, 2002). Previous studies showed an impaired autonomic nervous system functioning characterized by a limited heart rate variability in preterm-born infants as well as adolescents (Haraldsdottir et al., 2018; Patural et al., 2004). Preterms, however, showed differential SCRs to the CS + and CS-.

Although very preterm-born young adults were on average more anxious than their term-born peers, we did not detect significant abnormalities in the acquisition of learned differential fear responses. Likewise, extinction learning and return of fear in recall and reinstatement were not significantly altered. Small differences, however, may become more obvious using a simpler differential learning paradigm with no change in contexts and a single CS+. Abnormalities in preterms may also be more prominent in preterms who have developed more pronounced anxiety disorders. Findings need to be confirmed in future studies in larger preterm populations and using less complex fear conditioning paradigms.

Participants

A total of 38 preterm-born young adults and 42 age- and sex-matched term-born controls performed the experiment. Decision was made to exclude 12 participants (1 preterm participant and 11 controls) tested prior the COVID-19 pandemic and therefore without the necessary safety measures (face mask, transparent divider between examiner and participant) which likely changed the baseline level of fear. Thus, a total of 37 preterm-born participants (mean age 20.0 ± 2.8 years) and 31 controls (mean age 22.2 ± 2.3 years) were included in the final data analysis.

Inclusion criteria were as follows: i) very preterm birth (23–≤ 32 wks) or term birth (≥ 37 wks), and ii) age-appropriate development at the time of the testing without special needs/education. Exclusion criteria were: i) intra/ (peri-) ventricular hemorrhage ≥ III or periventricular leukomalacia based on brain MRI or ultrasound acquired at the time of term equivalent age, ii) focal neurological disorders. For clinical description of the included participants see Table 1.

Twelve preterm-born participants and one control had a history of emotional disorders. Five preterms developed (social) phobias and/or subclinical symptoms of anxiety or depression in their adolescence, without acute symptoms and without medication at the time of the experiment. Five preterm participants developed an attention deficit syndrome with or without hyperactivity in childhood/adolescence, without acute symptoms and without treatment at the time of the experiment. Two preterms reported difficulties in their social-emotional adaptation as a child and received occupational therapy. One control presented with a somatoform pain disorder in adolescence.

Fifteen preterm-born participants showed developmental disorders concerning language, gross- or fine motor functions which needed therapy in the past (speech therapy, physical therapy or occupational therapy). Thirtyfour very preterm-born participants and 30 controls were right-handed, two very preterm-born participants were left-handed and one very preterm-born participant and one control were ambidextrous based on the Edinburgh handedness inventory (Oldfield, 1971). Participants were instructed to refrain from alcohol and drug intake at least 24 hours prior to the experiment. All of the participants were non-smokers.

The ethics committee of the University of Duisburg-Essen approved the study (19-8890-BO). The study conforms to the principles laid down in the Declaration of Helsinki. All participants gave written informed consent. They were compensated for their participation with 80 Euros.

Depression-Anxiety-Stress-Scale-21 (DASS-21)

The Depression-Anxiety-Stress-Scale-21 (DASS-21) questionnaire was used to assess participants’ depression, anxiety, and stress levels (Fahrenberg, 2010). The DASS-21 is a 21-question self-report with 7 questions for each of the three subscales. On the depression subscale a score of 0–9 is within the normal range, on the anxiety subscale a score of 0–7, and on the stress subscale a score of 0–14 (Lovibond et al., 1995).

Fear conditioning

The experiment was performed on two consecutive days. Figure 1 displays the experimental paradigm. Habituation, acquisition training and extinction training was performed on day 1, recall and reinstatement were tested on day 2.

The experimental paradigm closely followed the paradigm introduced by Milad et al. (2007a). To emphasize conditioning to the cue and not the context, acquisition and extinction training were performed in different contexts (A and B) represented by pictures of two different office spaces. Conditioned stimuli (CS) were represented by a lamp shining either in blue, red or yellow color. The unconditioned stimulus (US) was an electric shock presented to the left calf. The same desk lamp was present in both contexts. Two CS + s and one CS- were shown in the acquisition training phase. The CS + s were both reinforced by an electric shock. Reinforcement rate was 66.6%. The CS- was never paired with the shock. In extinction training, one of the CS + s and the CS- were shown. In recall and reinstatement the extinguished (CS + E), the unextinguished (CS + U) and the CS- were shown. A CS + U was used in addition to a CS + E for direct comparison of recall of extinction (CS + E) and fear (CS + U) in early recall (Milad et al., 2007a).

The experimental protocol on day 1 consisted of three phases: "habituation" (2 CS + E only trials, 2 CS + U only trials, 2 CS only trials, presented in acquisition context A), "acquisition" (8 paired CS + E/US trials, 8 paired CS + U/US trials, 4 CS + E only trials, 4 CS + U only trial, 12 CS only trials, presented in acquisition context A) and "extinction" (12 CS + E only trials, 12 CS only trials, presented in extinction context B). The experimental protocol on day 2 consisted of the recall phase (8 CS + E only trials, 8 CS + U only trials, 16 CS only trials, presented in extinction context B), and the reinstatement phase (4 US-only trials in neutral background followed by 8 CS + E only trials, 8 CS + U only trials, 16 CS only trials, presented in extinction context B) phases. A checkerboard (with four two darker and two lighter gray squares) was shown as neutral background. The different trial types in each phase were presented in pseudo-randomized order, with two restrictions: Firstly, the first two trials and the very last trial of acquisition phase were set to be paired CS+/US trials, and secondly, the number of events of each kind was kept identical in the first half and in the second half of the experiment.

The order of events was the same for all participants in habituation and extinction phases. In the acquisition phase, the order of CS + E and CS + U events was counterbalanced. In the recall and reinstatement phases the first event was counterbalanced between CS + E and CS + U trials.

The paradigm presentation was controlled by a computer running the software Presentation (version 20.0, Neurobehavioral System Inc., Berkeley, CA). Visual stimuli were shown on a monitor (52.2 cm x 29.4 cm) placed about 6.5 feet (ca. 1.90 meter) infront of the participant. The context image was continously displayed throughout each phase. Each trial consisted of an 8 s CS presentation. In case of reinforced trials, a 100 ms aversive US was presented co-terminating with the CS+. Intertrial intervals were randomized between 12.2 s and 15.7 s. The use of context images and CS colors was pseudo-randomly counterbalanced across participants.

A short electrical stimulation was used as an aversive US. The electrical stimulation was generated by a constant current stimulator (DS7A, Digitimer Ltd., London, UK) and applied to the left calf (over the gastrocnemius muscle) via a concentric (ring-shaped) bipolar surface electrode with 6 mm conductive diameter and a central platinum pin (WASP electrode, Specialty Developments, Bexley, UK). Electrode position was marked with a permanent marker on day 1 to use the same electrode position on day 2. The 100 ms US consisted of a short train of four consecutive 500 µs current pulses (maximum output voltage: 400 V) with an interpulse interval of 33 ms. Stimulation intensity was determined immediately before start of the experiment. Stimulation current was gradually increased, and participants were asked to report on the perceived sensation intensity until an “unpleasant” but not “painful” intensity was reached. To counteract habituation to the unpleasant stimuli leading to weakening of the CRs, 20% was added to the individual thresholds (Inoue et al., 2020). The final individual current setting was kept constant for all stimulations.

On both days, immediately prior to the experiment each participant read instructions on the screen, that stated that they would be shown visual stimuli and that electrical shocks would be applied during the experiment. On day 1 participants were informed, that should they perceive a pattern between stimuli, the experimenter would not change that pattern during the experiment. On day 2, participants were informed that any pattern perceived during day 1 would stay the same on day 2. Participants confirmed that they had read and understood the instructions. Prior habituation on day 1, participants were informed that during the following phase only pictures will be shown without presentation of an electric shock.

Acquisition and evaluation of skin conductance responses

Throughout the experiment, skin conductance responses (SCRs) were measured (see Badsikadze et al., 2022 for details). SCRs were acquired using a physiological data acquisition station and appropriate hardware filters sampling at 1 kHz (MP160, BIOPAC Systems Inc., Goleta, CA). Two skin conductance electrodes were attached to the participant's left hypothenar, approximately 2 cm apart from each other.

To eliminate high-frequency noise SCR data was low-pass filtered with a 10 Hz cutoff using a hardware filter (EDA100C-MRI module, BIOPAC Systems Inc., Goleta, CA). Semi-automated peak detection was performed using MATLAB software (Release 2019a, RRID:SCR_001622, The MathWorks Inc., Natick, MA), and SCRs were defined as the maximum trough-to-peak-amplitude of any SCR peak starting within a time interval from 1 to 7.999 s after CS onset (Pineles et al., 2009).

The resulting raw SCR amplitudes were averaged in blocks and normalized through logarithmic (ln[1 + SCR]) transformation (Boucsein et al., 2012). Shapiro-Wilk-test was used to test the data and the distribution of residuals for normality. Since the normality test revealed a non-normal distribution of SCRs and the residuals (p < 0.05), data were analysed with non-parametric statistical analysis using the PROC Mixed procedure in SAS (SAS Studio 3.8, SAS Institute Inc, Cary, NC, USA). Non-parametric ANOVA-type statistics for repeated measures (Brunner et al., 2002; Shah et al., 2004) was used separately for each phase with SCR as dependent variable, stimulus (CS + E, CS + U, CS) and block (early and late phase) as within-subject factors and group (preterm, control) as between subject factor as well as their interactions. In case of significant results of non-parametric ANOVA, post-hoc comparisons were performed using least square means tests and were adjusted for multiple comparisons using the Tukey-Krammer method.

Questionnaires

Participants were asked to answer four questionnaires following each phase of the experiment (see Batsikadze et al., 2022, for details). Questions were shown on the monitor and participants gave answers using a button box with their right hand.

Participants were asked to rate their (hedonic) valence, (emotional) arousal, fear and contingency awareness on viewing images of the CS + E, CS + U and CS on a nine-step Likert scale from "very pleasant" to "very unpleasant", "quiet and relaxed" to "very excited", “not afraid” to “very afraid” and “US not expected” to “US surely expected”, respectively. Additionally, the questionnaire following acquisition contained further questions regarding US perception and CS-US contingency: rating of the last US on a nine Likert step-scale ("not unpleasant" to "very unpleasant"), and an estimation after which time and number of US presentations, if at all, a connection between the visual stimuli and the US presentation was identified.

For conditioning and CS-US contingency awareness assessment, the valence, arousal, fear, and US expectancy ratings were analyzed using non-parametric ANOVA type statistic for repeated measures with the respective rating as a dependent variable, stimulus (CS + E, CS + U, CS-) and time (prior acquisition, post acquisition, post extinction and post recall) as within-subject factors and group (preterm, control) as between subject factor, as well as their interactions. In case of the significant results of non-parametric ANOVA, post-hoc comparisons were performed using least square means tests and were adjusted for multiple comparisons using the Tukey-Krammer method.

Data availability statement

All data generated or analysed during this study are included in this published article and its supplementary information files.

Acknowledgements

This work was supported by UMEA (University Medicine Essen Clinician Scientist Academy) and a grant from the German Research Foundation (DFG; project number 316,803,389 – SFB 1280) to D.T. (subproject A05).

Agrawal, S., Rao, S. C., Bulsara, M. K., & Patole, S. K. (2018). Prevalence of Autism Spectrum Disorder in Preterm Infants: A Meta-analysis. Pediatrics, 142(3), e20180134. doi:10.1542/peds.2018-0134
Ancel, P. Y., Goffinet, F., Kuhn, P., Langer, B., Matis, J., Hernandorena, X., . . . Kaminski, M. (2015). Survival and morbidity of preterm children born at 22 through 34 weeks' gestation in France in 2011: results of the EPIPAGE-2 cohort study. JAMA Pediatr, 169(3), 230-238. doi:10.1001/jamapediatrics.2014.3351
Anderson, P. J., Cheong, J. L. Y., & Thompson, D. K. (2015). The predictive validity of neonatal MRI for neurodevelopmental outcome in very preterm children. Seminars in Perinatology, 39(2), 147-158. doi:https://doi.org/10.1053/j.semperi.2015.01.008
Batsikadze, G., Diekmann, N., Ernst, T. M., Klein, M., Maderwald, S., Deuschl, C., . . . Timmann, D. (2022). The cerebellum contributes to context-effects during fear extinction learning: A 7T fMRI study. Neuroimage, 253, 119080. doi:https://doi.org/10.1016/j.neuroimage.2022.119080
Bindt, C. (2022). Frühgeburt: Risiko für die psychische Gesundheit?: Wie elterliche Belastungen und frühkindliche Entwicklungsbedingungen zusammenwirken [Preterm birth: a risk for mental health?]. . Psychotherapeut 67(1), 28-33. doi:doi: 10.1007/s00278-021-00552-z.
Boucsein, W. (1992). Electrodermal Activity (2nd Edition ed.): Springer, New York.
Boucsein, W., Fowles, D. C., Grimnes, S., Ben-Shakhar, G., roth, W. T., Dawson, M. E., & Filion, D. L. (2012). Publication recommendations for electrodermal measurements. Psychophysiology, 49(8), 1017-1034. doi:10.1111/j.1469-8986.2012.01384.x
Bouton, M. E., & Bolles, R. C. (1979). Contextual control of the extinction of conditioned fear. Learning and Motivation, 10(4), 445-466. doi:https://doi.org/10.1016/0023-9690(79)90057-2
Bouton, M. E., & King, D. A. (1983). Contextual control of the extinction of conditioned fear: tests for the associative value of the context. J Exp Psychol Anim Behav Process, 9(3), 248-265.
Breeman, L. D., Jaekel, J., Baumann, N., Bartmann, P., & Wolke, D. (2016). Attention problems in very preterm children from childhood to adulthood: the Bavarian Longitudinal Study. J Child Psychol Psychiatry, 57(2), 132-140. doi:10.1111/jcpp.12456
Britton, J. C., Lissek, S., Grillon, C., Norcross, M. A., & Pine, D. S. (2011). Development of anxiety: the role of threat appraisal and fear learning. Depress Anxiety, 28(1), 5-17. doi:10.1002/da.20733
Brunner, E., Domhof, S., & Langer, F. (2002). Nonparametric analysis of longitudinal data in factorial experiments. New York, NY: J. Wiley.
Chau, C. M. Y., Ranger, M., Bichin, M., Park, M. T. M., Amaral, R. S. C., Chakravarty, M., . . . Grunau, R. E. (2019). Hippocampus, Amygdala, and Thalamus Volumes in Very Preterm Children at 8 Years: Neonatal Pain and Genetic Variation. Front Behav Neurosci, 13, 51. doi:10.3389/fnbeh.2019.00051
Chen, Q., Yang, H., Rooks, B., Anthony, M., Zhang, Z., Tadin, D., . . . Lin, F. V. (2020). Autonomic flexibility reflects learning and associated neuroplasticity in old age. Hum Brain Mapp, 41(13), 3608-3619. doi:10.1002/hbm.25034
Christie, M. J. (1981). Electrodermal activity in the 1980s: a review. J R Soc Med, 74(8), 616-622.
Cohen, R. T., & Kahana, M. J. (2022). A memory-based theory of emotional disorders. bioRxiv, 817486. doi:10.1101/817486
Craske, M. G., Waters, A. M., Lindsey Bergman, R., Naliboff, B., Lipp, O. V., Negoro, H., & Ornitz, E. M. (2008). Is aversive learning a marker of risk for anxiety disorders in children? Behav Res Ther, 46(8), 954-967. doi:10.1016/j.brat.2008.04.011
Craske, M. G., Wolitzky-Taylor, K. B., Mineka, S., Zinbarg, R., Waters, A. M., Vrshek-Schallhorn, S., . . . Ornitz, E. (2012). Elevated responding to safe conditions as a specific risk factor for anxiety versus depressive disorders: evidence from a longitudinal investigation. J Abnorm Psychol, 121(2), 315-324. doi:10.1037/a0025738
Critchley, H. D. (2002). Electrodermal responses: what happens in the brain. Neuroscientist, 8(2), 132-142. doi:10.1177/107385840200800209
D'Onofrio, B. M., Class, Q. A., Rickert, M. E., Larsson, H., Långström, N., & Lichtenstein, P. (2013). Preterm birth and mortality and morbidity: a population-based quasi-experimental study. JAMA psychiatry, 70(11), 1231-1240. doi:10.1001/jamapsychiatry.2013.2107
de Vries, L. S., van Haastert, I. C., Benders, M. J., & Groenendaal, F. (2011). Myth: cerebral palsy cannot be predicted by neonatal brain imaging. Semin Fetal Neonatal Med, 16(5), 279-287. doi:10.1016/j.siny.2011.04.004
De Zeeuw, C. I., & Yeo, C. H. (2005). Time and tide in cerebellar memory formation. Curr Opin Neurobiol, 15(6), 667-674. doi:10.1016/j.conb.2005.10.008
Duits, P., Baas, J. M. P., Engelhard, I. M., Richter, J., Huisman-van Dijk, H. M., Limberg-Thiesen, A., . . . Cath, D. C. (2021). Latent class growth analyses reveal overrepresentation of dysfunctional fear conditioning trajectories in patients with anxiety-related disorders compared to controls. J Anxiety Disord, 78, 102361. doi:10.1016/j.janxdis.2021.102361
Duits, P., Cath, D. C., Lissek, S., Hox, J. J., Hamm, A. O., Engelhard, I. M., . . . Baas, J. M. (2015). Updated meta-analysis of classical fear conditioning in the anxiety disorders. Depress Anxiety, 32(4), 239-253. doi:10.1002/da.22353
Duits, P., Richter, J., Baas, J. M. P., Engelhard, I. M., Limberg-Thiesen, A., Heitland, I., . . . Cath, D. C. (2017). Enhancing effects of contingency instructions on fear acquisition and extinction in anxiety disorders. J Abnorm Psychol, 126(4), 378-391. doi:10.1037/abn0000266
Ernst, T. M., Brol, A. E., Gratz, M., Ritter, C., Bingel, U., Schlamann, M., . . . Timmann, D. (2019). The cerebellum is involved in processing of predictions and prediction errors in a fear conditioning paradigm. Elife, 8. doi:10.7554/eLife.46831
Eryigit-Madzwamuse, S., Strauss, V., Baumann, N., Bartmann, P., & Wolke, D. (2015). Personality of adults who were born very preterm. Arch Dis Child Fetal Neonatal Ed, 100(6), F524-529. doi:10.1136/archdischild-2014-308007
Fahrenberg, J., Myrtek, M., Brähler, E., & Schumacher, J. (2010). Life Satisfaction Questionnaire. Primary data from the standardization sample 1994.
Frazier, J. A., Cochran, D., Kim, S., Jalnapurkar, I., Joseph, R. M., Hooper, S. R., . . . O'Shea, T. M. (2021). Psychiatric Outcomes, Functioning, and Participation in Extremely Low Gestational Age Newborns at Age 15 Years. J Am Acad Child Adolesc Psychiatry. doi:10.1016/j.jaac.2021.12.008
Gerwig, M., Dimitrova, A., Kolb, F. P., Maschke, M., Brol, B., Kunnel, A., . . . Timmann, D. (2003). Comparison of eyeblink conditioning in patients with superior and posterior inferior cerebellar lesions. Brain, 126(Pt 1), 71-94. doi:10.1093/brain/awg011
Gewirtz, J. C., Falls, W. A., & Davis, M. (1997). Normal conditioned inhibition and extinction of freezing and fear-potentiated startle following electrolytic lesions of medical prefrontal cortex in rats. Behav Neurosci, 111(4), 712-726. doi:10.1037//0735-7044.111.4.712
Gilchrist, C. P., Thompson, D. K., Alexander, B., Kelly, C. E., Treyvaud, K., Matthews, L. G., . . . Anderson, P. J. (2021). Growth of prefrontal and limbic brain regions and anxiety disorders in children born very preterm. Psychol Med, 1-12. doi:10.1017/s0033291721002105
Gustafsson, M., Josefsson, A., Ekholm Selling, K., & Sydsjö, G. (2009). Preterm birth or foetal growth impairment and psychiatric hospitalization in adolescence and early adulthood in a Swedish population-based birth cohort. Acta Psychiatr Scand, 119(1), 54-61. doi:10.1111/j.1600-0447.2008.01267.x
Haaker, J., Golkar, A., Hermans, D., & Lonsdorf, T. B. (2014). A review on human reinstatement studies: an overview and methodological challenges. Learn Mem, 21(9), 424-440. doi:10.1101/lm.036053.114
Haraldsdottir, K., Watson, A. M., Goss, K. N., Beshish, A. G., Pegelow, D. F., Palta, M., . . . Eldridge, M. W. (2018). Impaired autonomic function in adolescents born preterm. Physiol Rep, 6(6), e13620. doi:10.14814/phy2.13620
Inoue, L., Ernst, T. M., Ferber, II, Merz, C. J., Timmann, D., & Batsikadze, G. (2020). Interaction of Fear Conditioning with Eyeblink Conditioning Supports the Sensory Gating Hypothesis of the Amygdala in Men. eNeuro, 7(5), ENEURO.0128-0120.2020. doi:10.1523/ENEURO.0128-20.2020
Johnson, S., & Marlow, N. (2011). Preterm birth and childhood psychiatric disorders. Pediatr Res, 69(5 Pt 2), 11r-18r. doi:10.1203/PDR.0b013e318212faa0
Johnson, S., O'Reilly, H., Ni, Y., Wolke, D., & Marlow, N. (2019). Psychiatric Symptoms and Disorders in Extremely Preterm Young Adults at 19 Years of Age and Longitudinal Findings From Middle Childhood. J Am Acad Child Adolesc Psychiatry, 58(8), 820-826.e826. doi:10.1016/j.jaac.2019.02.020
Kim, M. J., Loucks, R. A., Palmer, A. L., Brown, A. C., Solomon, K. M., Marchante, A. N., & Whalen, P. J. (2011). The structural and functional connectivity of the amygdala: from normal emotion to pathological anxiety. Behav Brain Res, 223(2), 403-410. doi:10.1016/j.bbr.2011.04.025
Lærum, A. M., Reitan, S. K., Evensen, K. A., Lydersen, S., Brubakk, A. M., Skranes, J., & Indredavik, M. S. (2017). Psychiatric Disorders and General Functioning in Low Birth Weight Adults: A Longitudinal Study. Pediatrics, 139(2). doi:10.1542/peds.2016-2135
Lammertink, F., Vinkers, C. H., Tataranno, M. L., & Benders, M. J. N. L. (2021). Premature Birth and Developmental Programming: Mechanisms of Resilience and Vulnerability. Frontiers in psychiatry, 11, 531571-531571. doi:10.3389/fpsyt.2020.531571
Lattal, K. M., Radulovic, J., & Lukowiak, K. (2006). Extinction: [corrected] does it or doesn't it? The requirement of altered gene activity and new protein synthesis. Biol Psychiatry, 60(4), 344-351. doi:10.1016/j.biopsych.2006.05.038
Linnman, C., Zeidan, M. A., Furtak, S. C., Pitman, R. K., Quirk, G. J., & Milad, M. R. (2012). Resting amygdala and medial prefrontal metabolism predicts functional activation of the fear extinction circuit. Am J Psychiatry, 169(4), 415-423. doi:10.1176/appi.ajp.2011.10121780
Lissek, S., Powers, A. S., McClure, E. B., Phelps, E. A., Woldehawariat, G., Grillon, C., & Pine, D. S. (2005). Classical fear conditioning in the anxiety disorders: a meta-analysis. Behav Res Ther, 43(11), 1391-1424. doi:10.1016/j.brat.2004.10.007
Lissek, S., Rabin, S., Heller, R. E., Lukenbaugh, D., Geraci, M., Pine, D. S., & Grillon, C. (2010). Overgeneralization of conditioned fear as a pathogenic marker of panic disorder. Am J Psychiatry, 167(1), 47-55. doi:10.1176/appi.ajp.2009.09030410
Lissek, S., Rabin, S. J., McDowell, D. J., Dvir, S., Bradford, D. E., Geraci, M., . . . Grillon, C. (2009). Impaired discriminative fear-conditioning resulting from elevated fear responding to learned safety cues among individuals with panic disorder. Behaviour Research and Therapy, 47(2), 111-118. doi:https://doi.org/10.1016/j.brat.2008.10.017
Lonsdorf, T. B., Menz, M. M., Andreatta, M., Fullana, M. A., Golkar, A., Haaker, J., . . . Merz, C. J. (2017). Don't fear 'fear conditioning': Methodological considerations for the design and analysis of studies on human fear acquisition, extinction, and return of fear. Neurosci Biobehav Rev, 77, 247-285. doi:10.1016/j.neubiorev.2017.02.026
Lövdén, M., Bäckman, L., Lindenberger, U., Schaefer, S., & Schmiedek, F. (2010). A theoretical framework for the study of adult cognitive plasticity. Psychol Bull, 136(4), 659-676. doi:10.1037/a0020080
Lovibond, S. H., Lovibond, P. F., & Psychology Foundation of, A. (1995). Manual for the depression anxiety stress scales. Sydney, N.S.W.: Psychology Foundation of Australia.
Maschke, M., Schugens, M., Kindsvater, K., Drepper, J., Kolb, F. P., Diener, H. C., . . . Timmann, D. (2002). Fear conditioned changes of heart rate in patients with medial cerebellar lesions. J Neurol Neurosurg Psychiatry, 72(1), 116-118. doi:10.1136/jnnp.72.1.116
Matsuda, V. D. V., Tejada, M. B., Motta-Teixeira, L. C., Ikebara, J. M., Cardoso, D. S., Machado-Nils, A. V., . . . Takada, S. H. (2021). Impact of neonatal anoxia and hypothermic treatment on development and memory of rats. Exp Neurol, 340, 113691. doi:10.1016/j.expneurol.2021.113691
McCormick, D. A., & Thompson, R. F. (1984). Cerebellum: essential involvement in the classically conditioned eyelid response. Science, 223(4633), 296-299. doi:10.1126/science.6701513
Merz, C. J., Hamacher-Dang, T. C., & Wolf, O. T. (2016). Immediate extinction promotes the return of fear. Neurobiol Learn Mem, 131, 109-116. doi:10.1016/j.nlm.2016.03.013
Messerschmidt, A., Brugger, P. C., Boltshauser, E., Zoder, G., Sterniste, W., Birnbacher, R., & Prayer, D. (2005). Disruption of cerebellar development: potential complication of extreme prematurity. AJNR. American journal of neuroradiology, 26(7), 1659-1667. Retrieved from https://pubmed.ncbi.nlm.nih.gov/16091510
Milad, M. R., Wright, C. I., Orr, S. P., Pitman, R. K., Quirk, G. J., & Rauch, S. L. (2007a). Recall of fear extinction in humans activates the ventromedial prefrontal cortex and hippocampus in concert. Biol Psychiatry, 62(5), 446-454. doi:10.1016/j.biopsych.2006.10.011
Milad, M. R., Quirk, G. J., Pitman, R. K., Orr, S. P., Fischl, B., & Rauch, S. L. (2007b). A role for the human dorsal anterior cingulate cortex in fear expression. Biol Psychiatry, 62(10), 1191-1194. doi:10.1016/j.biopsych.2007.04.032
Murray, A. L., Scratch, S. E., Thompson, D. K., Inder, T. E., Doyle, L. W., Anderson, J. F., & Anderson, P. J. (2014). Neonatal brain pathology predicts adverse attention and processing speed outcomes in very preterm and/or very low birth weight children. Neuropsychology, 28(4), 552-562. doi:10.1037/neu0000071
Oldfield, R. C. (1971). The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia, 9(1), 97-113. doi:10.1016/0028-3932(71)90067-4
Omizzolo, C., Scratch, S. E., Stargatt, R., Kidokoro, H., Thompson, D. K., Lee, K. J., . . . Anderson, P. J. (2014). Neonatal brain abnormalities and memory and learning outcomes at 7 years in children born very preterm. Memory, 22(6), 605-615. doi:10.1080/09658211.2013.809765
Ortinau, C., & Neil, J. (2015). The neuroanatomy of prematurity: normal brain development and the impact of preterm birth. Clin Anat, 28(2), 168-183. doi:10.1002/ca.22430
Patural, H., Barthelemy, J. C., Pichot, V., Mazzocchi, C., Teyssier, G., Damon, G., & Roche, F. (2004). Birth prematurity determines prolonged autonomic nervous system immaturity. Clin Auton Res, 14(6), 391-395. doi:10.1007/s10286-004-0216-9
Pavlov, P. I. (2010). Conditioned reflexes: An investigation of the physiological activity of the cerebral cortex. Ann Neurosci, 17(3), 136-141. doi:10.5214/ans.0972-7531.1017309
Pesonen, A. K., Räikkönen, K., Heinonen, K., Andersson, S., Hovi, P., Järvenpää, A. L., . . . Kajantie, E. (2008). Personality of young adults born prematurely: the Helsinki study of very low birth weight adults. J Child Psychol Psychiatry, 49(6), 609-617. doi:10.1111/j.1469-7610.2007.01874.x
Pierre, W. C., Zhang, E., Londono, I., De Leener, B., Lesage, F., & Lodygensky, G. A. (2022). Non-invasive in vivo MRI detects long-term microstructural brain alterations related to learning and memory impairments in a model of inflammation-induced white matter injury. Behav Brain Res, 428, 113884. doi:10.1016/j.bbr.2022.113884
Pineles, S. L., Orr, M. R., & Orr, S. P. (2009). An alternative scoring method for skin conductance responding in a differential fear conditioning paradigm with a long-duration conditioned stimulus. Psychophysiology, 46(5), 984-995. doi:10.1111/j.1469-8986.2009.00852.x
Platt, M. J., Cans, C., Johnson, A., Surman, G., Topp, M., Torrioli, M. G., & Krageloh-Mann, I. (2007). Trends in cerebral palsy among infants of very low birthweight (<1500 g) or born prematurely (<32 weeks) in 16 European centres: a database study. The Lancet, 369(9555), 43-50. doi:10.1016/S0140-6736(07)60030-0
Rescorla, R. A., & Heth, C. D. (1975). Reinstatement of fear to an extinguished conditioned stimulus. J Exp Psychol Anim Behav Process, 1(1), 88-96.
Robbins, S. J. (1990). Mechanisms underlying spontaneous recovery in autoshaping. . Journal of Experimental Psychology: Animal Behavior Processes, 16(3), 235–249. doi:https://doi.org/10.1037/0097-7403.16.3.235
Sacchetti, B., Scelfo, B., & Strata, P. (2005). The cerebellum: synaptic changes and fear conditioning. Neuroscientist, 11(3), 217-227. doi:10.1177/1073858405276428
Schmitz-Koep, B., Zimmermann, J., Menegaux, A., Nuttall, R., Bäuml, J. G., Schneider, S. C., . . . Sorg, C. (2021). Decreased amygdala volume in adults after premature birth. Sci Rep, 11(1), 5403. doi:10.1038/s41598-021-84906-2
Shah, D. A., & Madden, L. V. (2004). Nonparametric analysis of ordinal data in designed factorial experiments. Phytopathology, 94(1), 33-43. doi:10.1094/PHYTO.2004.94.1.33
Steinhausen, H. C. (2013). [What happens to children and adolescents with mental disorders? Findings from long-term outcome research]. Z Kinder Jugendpsychiatr Psychother, 41(6), 419-431. doi:10.1024/1422-4917/a000258
Tam, E. W. Y. (2013). Potential mechanisms of cerebellar hypoplasia in prematurity. Neuroradiology, 55(2), 41-46. doi:10.1007/s00234-013-1230-1
Tran, L., Huening, B. M., Kaiser, O., Schweiger, B., Sirin, S., Quick, H. H., . . . Timmann, D. (2017). Cerebellar-dependent associative learning is impaired in very preterm born children and young adults. Scientific Reports, 7(1), 18028. doi:10.1038/s41598-017-18316-8
Van Lieshout, R. J., Boyle, M. H., Saigal, S., Morrison, K., & Schmidt, L. A. (2015). Mental health of extremely low birth weight survivors in their 30s. Pediatrics, 135(3), 452-459. doi:10.1542/peds.2014-3143
Volpe, J. J. (2009a). The Encephalopathy of Prematurity—Brain Injury and Impaired Brain Development Inextricably Intertwined. Seminars in Pediatric Neurology, 16(4), 167-178. doi:https://doi.org/10.1016/j.spen.2009.09.005
Volpe, J. J. (2009b). Brain injury in premature infants: a complex amalgam of destructive and developmental disturbances. The Lancet Neurology, 8(1), 110-124. doi:10.1016/S1474-4422(08)70294-1
Volpe, J. J. (2009c). Cerebellum of the premature infant: rapidly developing, vulnerable, clinically important. J Child Neurol, 24(9), 1085-1104. doi:10.1177/0883073809338067
Waters, A. M., Henry, J., & Neumann, D. L. (2009). Aversive Pavlovian conditioning in childhood anxiety disorders: impaired response inhibition and resistance to extinction. Journal of Abnormal Psychology 118(2), 311-321. doi:https://doi.org/10.1037/a0015635
Xia, D., Min, C., Chen, Y., Ling, R., Chen, M., & Li, X. (2020). Repetitive Pain in Neonatal Male Rats Impairs Hippocampus-Dependent Fear Memory Later in Life. Front Neurosci, 14, 722. doi:10.3389/fnins.2020.00722

No competing interests reported.

SupplementarydataScientificReportsFINAL.docx

Download PDF

Journal Publication

published 12 Jul, 2023

Read the published version in Scientific Reports →

Editorial decision: Major revision
27 Apr, 2023
Reviews received at journal
20 Apr, 2023
Reviews received at journal
01 Apr, 2023
Reviewers agreed at journal
21 Mar, 2023
Reviewers agreed at journal
08 Mar, 2023
Reviewers invited by journal
05 Mar, 2023
Editor assigned by journal
05 Mar, 2023
Editor invited by journal
02 Mar, 2023
Submission checks completed at journal
02 Mar, 2023
First submitted to journal
24 Feb, 2023

You are reading this latest preprint version

Fear conditioning is preserved in very preterm-born young adults despite increased anxiety levels

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Depression-Anxiety-Stress-Scale-21 (DASS-21) questionnaire

Skin conductance responses

Questionnaires

Comparing anxious and non-anxious preterm participants

Discussion

Conclusions

Materials & Methods

Participants

Depression-Anxiety-Stress-Scale-21 (DASS-21)

Fear conditioning

Acquisition and evaluation of skin conductance responses

Questionnaires

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1