Latent inhibition is when prior benign experience with a cue can impede or inhibit subsequent emotional learning to the cue (Lubow, 1973; Mineka and Zinbarg, 2006). In Pavlovian conditioning, latent inhibition is observed when the pre-exposure to the conditioned stimulus (CS) interferes with acquisition (George et al., 2010; Shalev et al., 1998; Lubow and Moore, 1959) or retrieval (Leung et al., 2013; Lingawi et al., 2016) of CS association with the unconditioned stimulus, measured by decreased conditioned responding in pre-exposed compared to non-pre-exposed subjects. Latent inhibition is proposed to explain how prior benign experiences prevent the development of experience-based anxiety disorders (Mineka and Zinbarg, 2006). For example, children who did not initially experience stress with dentists are much less likely to develop dental phobias following a subsequent traumatic experience (Davey, 1989). Similarly, monkeys who have observed models playing with snakes in seemingly safe circumstances do not develop fearful behaviors when directly exposed to snakes (Mineka and Cook, 1986).

Adolescence is a period of elevated risk to anxiety disorders (Polanczyk et al., 2015), which may be associated with latent inhibition. Indeed, anxiety symptoms in humans are related to deficits in inhibitory learning and memory measured by extinction of conditioned fear (Ganella et al., 2018a; Marin et al., 2017). Like latent inhibition, extinction involves benign exposures to a cue, except that such exposure occurs after the emotional learning (Bouton, 1993). Adolescents have been shown to display extinction impairments compared to adults in rats and humans (Ganella et al., 2017a,2018a; Ganella and Kim, 2014; Pattwell et al., 2012, 2013; Zbukvic et al., 2017). One study has shown that postnatal day 35 (P35) adolescent male rats indeed show latent inhibition, as measured by conditioned suppression of licking behavior in response to a tone CS that was paired with shocks (Zuckerman et al., 2003). However, latent inhibition has not yet been directly compared between adolescents and adults, which is the first aim of the present study. We hypothesize that the behavior of adolescents and adults will be differentially affected by latent inhibition because vulnerability towards anxiety disorders in adolescence (McGrath et al., 2023) has long been associated with the reduced function of medial prefrontal cortex (mPFC), which undergoes dramatic changes in structure and neurochemistry across adolescence (Casey et al., 2008; Kim et al., 2017; Perry et al., 2021; Gold et al., 2020). Notably, previous literature in rats gives rise to opposing predictions. On the one hand, temporary inactivation of the infralimbic cortex (IL) of mPFC led to deficits in latent inhibition (Lingawi et al., 2016, 2018), which suggests that latent inhibition would be impaired in adolescence. On the other, IL lesions facilitated latent inhibition (George et al., 2010), which suggests that latent inhibition would be facilitated in adolescents. Moreover, mPFC lesions did not affect latent inhibition in other studies (Joel et al., 1997; Lacroix et al., 1998) suggesting that latent inhibition may be comparable in adolescent and adult rats.

The second aim of the present study is to assess latent inhibition in males and females. Biological sex is a significant factor for anxiety disorders, with greater prevalence reported within the female population (Kessler et al., 2005). Sex differences in inhibitory learning and memory have been observed (Milad et al., 2009), although this effect is strongly mediated by hormonal effects/estrous cycle in female rodents and humans (Lebron-Milad et al., 2012; Graham and Milad, 2013). Interestingly, these estrous cycle effects are age dependent. Stages where there are high levels of estradiol are associated with improved extinction in adult females (Lebron-Milad et al., 2012; Graham and Milad, 2013), and impaired extinction in adolescent females (Perry et al., 2020). In addition, sex effects in extinction have been associated with differences in mPFC function (Day et al., 2020). Hence, latent inhibition may also be sex-specific. To our knowledge, there have been no studies that have examined latent inhibition of conditioned fear in female rodents. We hypothesize females will show age-specific effects of estrous cycle on latent inhibition. Specifically, latent inhibition would be greater when adult rats are in proestrous (i.e., when estradiol levels are high), while the opposite may be true in adolescent rats.

The aim of this study was to examine age differences (adolescent vs. adult) in latent inhibition in male and female rats. We found that adolescents were more likely to display evidence of latent inhibition when compared to adults, with lower levels of freezing shown in adolescents compared to adults in the pre-exposed rats. These findings do not exclude that latent inhibition occurs for adults as well, since there is evidence of delayed conditioning in the pre-exposed rats on day 2 regardless of age and sex (Figures 1, 3). This delayed learning was not due to decreased sensitivity to the shock as a result of pre-exposure to the CS. If anything, pre-exposed animals showed greater sensitivity, although this effect was only evident in adolescent males. Thus, during pre-exposure both adults and adolescents had learned that the CS has no aversive consequence, and this learning delays acquisition of the CS-shock association similarly across age groups. However, the effect of the initial experience is clearly more enduring for adolescents, since pre-exposure was more likely to reduce fear expression on day 3 in that age group. Surprisingly, despite previous studies that fear learning is affected by sex hormones (Graham and Milad, 2013, 2014; Perry et al., 2020), we found no effect of estrous in our female rats.

The enhanced latent inhibition observed here in adolescents is similar to effects seen following experimental lesion of the IL (George et al., 2010). Compared to other brain regions, the mPFC is last to reach maturity in rodents and humans (Casey et al., 2008; Kim et al., 2017; Gogtay et al., 2004; Giedd, 2004; Bjerke et al., 2025). It is therefore possible that the enhanced latent inhibition effect observed here in adolescents may be due to reduced functionality in the IL mPFC, which led to stronger effects of pre-exposure. When IL lesions facilitated latent inhibition, George et al. (2010) argued that the role of IL is to allow the learning of a second, conflicting information about a cue. As such, IL hypofunction during adolescence results in enhanced latent inhibition because the second association (cue-shock) is not encoded properly. Such a role for IL is also consistent with previous studies showing impaired extinction and reduced IL function in adolescents (Ganella et al., 2018a; Kim et al., 2011). In contrast, others have shown that temporary inactivation of the IL inhibited retrieval of latent inhibition (Lingawi et al., 2016, 2018). The differences in findings may be explained by specific methodology. In George et al. (2010), IL lesions occurred prior to any behavioral training. Thus, function was impaired across pre-exposure and at test. When IL inactivation only occurred immediately prior to retrieval test (Lingawi et al., 2016, 2018), IL function would have been intact during pre- exposure. From this, IL may be critical for remembering the latent inhibition memory once pre-exposure and conditioning have already occurred. In our study, assumed IL hypofrontality was due to developmental stage rather than experimental manipulation, hence would have been present during pre-exposure and conditioning as with George et al. (2010).

Importantly, brain maturation is not restricted to the IL and the present results may be due to dopamine signaling in the nucleus accumbens (NAc). A recent study showed that cue-elicited dopamine release in the NAc core was directly linked to latent inhibition (Kutlu et al., 2022). Specifically, exposure to the cue evokes dopamine release, which declines with repeated exposures across the pre-exposure session. Therefore, during conditioning, the pre-exposed cue evokes a smaller dopamine release than the non-pre-exposed cue, meaning that it is less able to form an association with the US (Kutlu et al., 2022). At conditioning, intra-nucleus accumbens (NAc) injections of amphetamine attenuated, while a dopamine receptor 2 (D2) antagonist (haloperidol) enhanced latent inhibition (Joseph et al., 2000). However, intra-mPFC injection of amphetamine or apomorphine had no effects (Lacroix et al., 2000; Broersen et al., 1999). Furthermore, deficits in latent inhibition of conditioned fear due to adolescent social isolation were associated with increased D2 expression in the mPFC and NAc (Han et al., 2012). Conversely, a dopamine receptor 1 (D1) agonist administered systemically prior to pre-exposure and conditioning sessions mildly inhibited latent inhibition of fear conditioning (Feldon et al., 1991). A D1 antagonist (SCH23390) administered alone had no effect, but did block amphetamine-induced disruption of latent inhibition (Nelson et al., 2012). In summary, attenuated D1 or D2 signaling appears to promote latent inhibition, which can explain the present age differences observed with D1 reduced in the ventral striatum but not mPFC in adolescence compared to adulthood (Bjerke et al., 2025; Cullity et al., 2019). Future studies should look at differential effects of D1 and D2 receptor modulators in adults and adolescents in order to better understand these effects.

Latent inhibition involves targeted redirection of attention following pre-exposure to the stimulus (Colagiuri et al., 2021; Lubow, 2005). In other words, there is decreased attention directed towards to the target stimulus following pre-exposure, making it less available for association to the US in subsequent conditioning sessions. Children with attention deficit hyperactivity disorder showed reduced latent inhibition compared to controls in a discrimination task (Lubow and Josman, 1993) and a visual search task (Lubow, 2005). Within this context, the current findings suggest that adolescents were more able to attend to non-predictive stimuli than adults. However, we also observed that both adults and adolescents acquired conditioned fear to the non-pre-exposed stimulus, meaning that both ages are equally able to attend to a stimulus when it is novel. Further, the age difference in latent inhibition only emerged at test, which suggests a retrieval rather than an encoding effect. Nevertheless, this does not rule out that a weaker association was formed between the pre-exposed CS and the US due to increased attention in the adolescent group.

We did not find any effects for estrous cycle during conditioning or at test. In contrast, extinction is impacted by the estrous phase (Milad et al., 2009; Perry et al., 2020), and by the experimental manipulation of gonadal hormones (Perry et al., 2020; Zeidan et al., 2011; Maeng et al., 2017). Interestingly, this effect on extinction is also apparently age-dependent, because in adults high levels of estrogen facilitate extinction (Milad et al., 2009; Maeng and Milad, 2015; Velasco et al., 2019), while in adolescents extinction is facilitated where estrogen is not present/present at low levels (Perry et al., 2020). In the current study there were no differences in latent inhibition reported where rats were at different stages of estrous cycle, regardless of age. This suggests that interference between a benign memory and a traumatic experience is not dependent on cycling gonadal hormones and implies that there are distinct neural mechanisms underlying extinction and latent inhibition. Notably, the present study observed overall reduced variation in estrous cycling in adolescents due to many animals staying unopened throughout the behavioral days, which is different from the previous study in which the adolescent female rats of the same age were much more variable in their estrus phases (Perry et al., 2020). It is possible that testing adults and adolescents together as in the present study affected estrous cycling, as previous studies have shown female rats in proximity can affect their estrous phase (Alekhina et al., 2015). Different outcomes may be observed if the adolescent and adult females are not concurrently tested.

A limitation of this study is that male and female cohorts were run in separate experiments, meaning that direct statements regarding sex differences cannot be made. We chose to treat the two cohorts as separate experiments due to differences in shock sensitivity. Specifically, male rats showed robust freezing and latent inhibition effect (at least for adolescents) using a 0.45 mA shock, while for females freezing was unreliable using these parameters, and the shock needed to be increased to 0.6 mA in order to obtain reliable conditioning. Sex differences in sensitivity to footshock and consequent conditioning have been observed previously (Stock et al., 2001; Gupta et al., 2001; Wiltgen et al., 2001; Baran et al., 2009), although this is not always the case (Orsini et al., 2016). Sex differences in latent inhibition have been reported for a conditioned taste aversion whereby adult female rats showed weaker latent inhibition than males (Angulo et al., 2020; Angulo and Arévalo-Romero, 2021). Interestingly, the same study saw delayed extinction in males compared with females (Angulo and Arévalo-Romero, 2021), which is different from what we (Perry et al., 2020) and others (Velasco et al., 2019) have observed. Furthermore, human participants did not show sex differences in latent inhibition for a discrimination task (Lubow and Josman, 1993). Indeed, it is possible that sex specificity of latent inhibition is task-dependent.

Notably, all sessions occurred in the same context in the present study, which means that latent inhibition of context conditioning may have occurred (Killcross et al., 1998; Westbrook et al., 1994; Kiernan and Westbrook, 1993). That is, prolonged pre-exposure to a context can reduce subsequent conditioning in that context. However, typical latent inhibition of contextual conditioned fear involves extensive pre-exposure to the context, such as 20 min twice a day for 11 days (Killcross et al., 1998). Interestingly, shorter exposure to the context can lead to “context pre-exposure facilitation effect” (Heroux et al., 2018; Fanselow, 1986; Robinson-Drummer and Stanton, 2015), which is increased context conditioned fear when there is a presentation of the footshock as soon as the rodent is placed into the context. Without context pre-exposure, such immediate shock in a context leads to a failure in context conditioning (“immediate shock deficit”) (Fanselow, 1986; Robinson-Drummer and Stanton, 2015). Such deficit is not so apparent with discrete cues, highlighting how forming a contextual representation requires longer time than forming a discrete cue representation (Rudy, 1993; Kim and Richardson, 2009). This may be due to the necessity of the different elements in the context to come together to form a spatial representation in contextual learning (Rudy and Sutherland, 1995). In any case, with the inclusion of “No pre-exposure” group that received an identical amount of context exposure as the pre-exposed group, it is unlikely that the present age effects are due to differences in latent inhibition of contextual conditioned fear.

In summary, the current findings show that adolescents show stronger latent inhibition than adults. Latent inhibition serves an adaptive purpose (Lubow, 2005). For example, it helps to focus attention away from stimuli less likely to be relevant in a new situation, thereby preserving cognitive load. It may also prevent too much predictive value being attributed to stimuli that are only salient in passing/in certain contexts. Thus, this facilitated latent inhibition in adolescents reflects experience-dependent resilience as they are able to form robust CS-no-event association that is more readily retrieved following conditioning when the rats are tested in extinction (i.e., safe) conditions (Mineka and Zinbarg, 2006). The fact that age-differences in expression of latent inhibition did not emerge until test (under extinction conditions) provides weight to this interpretation. However, future studies should explore this possibility among others, such as reduced behavioral flexibility and changes in attention, in order to provide a more comprehensive understanding on how learning styles change across development. Anxiety disorders often emerge in adolescence, and this has led to the perspective that adolescents are more vulnerable (Polanczyk et al., 2015; Ganella et al., 2018a; Kim and Ganella, 2015). The current findings show that this story is more complex, because facilitated latent inhibition may provide resilience against development of anxiety. Notably, this resilience is dependent on experiential factors, since latent inhibition is dependent on prior exposure to the stimuli. In other words, timing of traumatic experiences relative to other experiences across adolescence is critical for determining how much of an impact the trauma will have on subsequent behavior.