Negative (especially highly arousing) emotional stimuli tend to be retained longer in memory and retrieved more accurately than positive or neutral ones, as remembering negative information increases the chance of survival due to its prioritized evolutionary value (Kensinger & Schacter, 2016; see also Mather & Sutherland, 2009 for an arousal-based, not valenced-based, account of emotional memory.) Recent studies, however, have demonstrated that specific negative emotions can differentially influence recognition memory performance. In particular, disgust-related items have consistently been shown to have a robust advantage in recognition accuracy compared to fear- or anger-related items (e.g., Boğa et al., 2021, Chapman, 2018, Chapman et al., 2013, Davis et al., 2011, Ferré et al., 2018, Kapucu et al., 2024, Marchewka et al., 2016, Wang, 2021; see also Chapman, 2021, West and Mulligan, 2021 Exp1, Exp2 for null effects of a disgust-related memory advantage), despite having similar valence (i.e., how pleasant or unpleasant an emotion is) and arousal levels (i.e., how exciting or calming an emotion is). Thus, in contrast to what early emotional memory research claimed (for a review, see Kensinger & Schacter, 2016), valence or arousal levels cannot be the primary determinants of the emotion effects on memory, suggesting that the effects of emotions on memory should be investigated from a more nuanced perspective than the basic dimensional approach, by incorporating the idiosyncratic qualities and appraisal properties of specific emotions. Furthermore, the underlying decision processes and biases during retrieval that drive this disgust-related advantage have yet to be clearly understood. In the present study, within the framework of appraisal theories of emotion (Lerner and Keltner, 2000, Smith and Ellsworth, 1985), as an alternative approach to the valence/arousal dimensions, we aimed to investigate the effects of disgust and fear on recognition memory, utilizing the diffusion model of recognition memory (Ratcliff, 1978, Ratcliff and McKoon, 2008) to uncover the decision process and biases underlying memory performance.

Disgust and fear are the two specific negative emotions whose effects have often been contrasted to elucidate how specific emotions shape recognition memory performance. The rationale behind contrasting these two emotions in particular is that both are negative and highly arousing (Russell, 1980). Since both emotions signal an environmental threat (e.g., the threat of an aggressive animal or pathogen), they evoke motivational tendencies to facilitate avoidance behavior (Krusemark & Li, 2011). Thus, these two emotions are well-suited for researchers aiming to control for the potential confounding effects of valence, arousal, and motivational tendency, and focus solely on the effects of specific emotions on memory processes.

The robust accuracy advantage of disgust-related items over fear-related or neutral ones has been demonstrated in numerous studies using different learning methods (incidental, West & Mulligan, 2021; intentional, Schienle et al., 2021), types of memory tasks (free recall, Chapman, 2018, Moeck et al., 2021; recognition memory, Boğa et al., 2021, Chapman et al., 2013), stimulus types (pictures, Boğa et al., 2021; words, Ferré et al., 2018, Charash and McKay, 2002, Riegel et al., 2022; faces, Zhang et al., 2017), age groups (children, Schienle et al., 2021; young adults, Chapman, 2018; older adults, Boğa et al., 2021), and even in the presence of concurrent secondary tasks during encoding (e.g., Chapman et al., 2013, Moeck et al., 2021, West and Mulligan, 2021). These findings suggest that disgust-related memory enhancement occurs consistently, usually regardless of the methodological characteristics of the study. Disgust, which serves primarily to protect the organism from contamination by potentially dangerous objects, seem to drive memory processes to retain evolutionarily privileged information (Carretié et al., 2011, Bradshaw and Gassen, 2021). Indeed, given the covert and easily transmissible nature of pathogens, experiencing disgust in response to pathogen-containing and infectious objects may increase the importance of retaining prior learning to avoid missing these covert threats in future encounters (Curtis et al., 2011), thereby enhancing the accuracy of memory judgments.

Research has focused on the different stages of memory to uncover the underlying mechanisms of the disgust-related memory advantage (e.g., Chapman et al., 2013, Moeck et al., 2021). Indeed, the question of which memory stages contribute most to emotional memory enhancement has long been investigated (LaBar and Cabeza, 2006, Phelps, 2004). Emotional (especially negative and highly-arousing) items have been argued to benefit from an encoding-based enhancement in memory due to prioritizing the attentional mechanisms during encoding (Phelps, 2004). On the other hand, special consolidation of emotional stimuli due to post-encoding interactions between the amygdala and hippocampus also plays an important role in emotional memory enhancement (McGaugh, 2004). Building on these accounts mostly limited to the valence/arousal dimensions, recent research has examined the role of encoding vs. consolidation-based mechanisms in the disgust-related memory advantage (e.g, Moeck et al., 2021). Eye-tracking studies demonstrated that disgust-related stimuli attracted more attention in the early stages of stimulus presentation compared to fear-related ones, suggesting that attentional processes may be prioritized at the encoding stage for disgust-related items (Calvo and Lang, 2004, Carretié et al., 2011, Van Hooff et al., 2013). Consistently, several studies have reported that the secondary task (i.e., line discrimination task) during encoding was performed more slowly for disgust-related stimuli than for fear-related stimuli, indicating an increased demand on attentional resources imposed by disgust-related stimuli (Chapman et al., 2013, Moeck et al., 2021). Nevertheless, although attention during the encoding phase appeared to mediate recall performance, the effects of disgust could not be fully explained by attentional processes alone. Boğa et al. (2021) also reported that the disgust-related advantage could not be explained by attentional resources as there was no disgust-specific slowing in the secondary task. Based on similar findings, West and Mulligan (2021) suggested that attentional bias potentially amplifies the disgust-related advantage when it is present, yet this advantage still emerges even in its absence. Alternatively, the evidence for disgust-specific memory consolidation appears to be more robust than that for attentional mechanisms. The disgust-related advantage over fear-related or neutral items tends to be stronger when participants’ memory was tested after a retention interval that allows for consolidation (Chapman et al., 2013, Moeck et al., 2021). Importantly, Moeck et al. (2021) argued that previous findings on the mediating role of attention during encoding may be due to the lack of statistical power and they contrasted both encoding/attention-based and consolidation-based accounts in the same research design. They reported that participants paid more attention to disgust- than to fear-related images during encoding. However, their mixed-effect models showed that disgust-specific consolidation, but not increased attention, might contribute to enhanced memory for disgust (Moeck et al., 2021). The disgust-specific brain region insula, which is involved in the consolidation process along with the amygdala, was argued to provide the physiological basis for these effects, due to its direct connections to the hippocampus (Calder et al., 2000, Vytal and Hamann, 2010). Yet, studies that focus on the disgust-related accuracy advantage, regardless of the memory stage at which this advantage occurs, can be considered theoretically and methodologically limited in several ways.

From a methodological perspective, measures of accuracy could be confounded by participants’ response biases in many previous studies, especially given that the liberal response bias (i.e., willingness to make “Old” recognition judgments regardless of the actual item type) is commonly observed with emotional stimuli (e.g., Dougal and Rotello, 2007, Grider and Malmberg, 2008). Indeed, model-based analyses revealed that emotional items created a subjective sense of recollection that led to a willingness to say “Old/Studied” by increasing misleading familiarity information, rather than actual use of the recollective process or retrieval of episodic contextual details. Yet, the potential confound of response bias has only been addressed in a few studies contrasting disgust- and fear-related items (e.g., Boğa et al., 2021, Chapman et al., 2013, Kapucu et al., 2024). Research utilizing the Signal Detection Theory (SDT, Macmillan & Creelman, 2005) to measure the separate effects of specific negative emotions on memory accuracy and response bias has revealed a different pattern of disgust-related advantage. To clarify, besides the accuracy advantage, disgust-related items also increased participants’ liberal response bias relative to fear-related or neutral items (Chapman et al., 2013, Kapucu et al., 2024, Exp2). On the other hand, fear-related items also increased participants’ liberal response bias relative to neutral ones, supporting the well-documented liberal response bias effects for negative emotional items compared to neutral ones (e.g., Dougal and Rotello, 2007, White et al., 2014, Windmann and Kutas, 2001; see also Grider & Malmberg, 2008 for the results supporting greater liberal response bias for positive emotional items), but no accompanying accuracy advantage was observed for fear-related items especially when accuracy was measured without response bias confound (Boğa et al., 2021, Chapman et al., 2013, Kapucu et al., 2024). That is, the liberal response bias effect for negative emotional items tends to be more pronounced when the emotional items carry disgust-related information, suggesting a disgust-specific increase in the subjective sense of recollection due to increased familiarity information. We note that the disgust-specific response bias effects may be less robust than the accuracy effects, as some studies found no disgust-specific increase in liberal response bias compared to other negative specific emotion conditions (Boğa et al., 2021, Kapucu et al., 2024, Exp1). Given the paucity of studies examining response bias effects, the effects of specific emotions on response bias have yet to be fully understood. More importantly, even when differences in response bias were addressed, the primary measures of accuracy in some studies (e.g., corrected recognition scores, Ferré et al., 2018) did not provide an independent estimate of memory accuracy and were confounded by differences in bias across conditions.

From a theoretical perspective, most emotional memory studies ignored the appraisal properties of specific emotions. (Un)certainty appraisal is a core appraisal among a set of appraisal dimensions that distinguish one emotion from another (see Smith & Ellsworth, 1985). (Un)certainty refers to what extent individuals feel uncertain about the actual source of the emotion and unsure about what will happen next (Lerner & Keltner, 2000). Based on appraisal theories of emotion (Lerner and Keltner, 2000, Smith and Ellsworth, 1985), disgust and fear differ in their levels of emotional (un)certainty despite sharing similar valence/arousal levels and motivational tendencies.

A central tenet of appraisal theories is that they conceptualize emotion as a process initiated by changes in the appraisal of the stimuli. Whereas the relation between stimuli and emotions may vary depending on the appraisal process, the relation between appraisals and emotions is assumed to be stable (Moors et al., 2013, p.121). Thanks to the variable relation between stimuli and emotions through the appraisal process, appraisal theories account for the diversity of emotional experiences beyond basic emotions. More importantly, due to the stable relation between appraisals and emotions, each specific emotion (e.g., fear) is considered to result from a consistent appraisal pattern. For example, fear does not arise when a stimulus is appraised as certain; rather, individuals experience fear only when the stimulus is evaluated with uncertainty and accompanied by a sense of lack of control. Therefore, appraisal and basic emotion approaches are not mutually exclusive (Ellsworth, 2024). Instead, appraisal theories may help to explain the underlying cognitive structure of basic emotions by linking them to specific situational evaluations.

According to appraisal theories, a high level of uncertainty increases the demand from our attentional resources as predicting the outcome of a decision is more challenging when feeling uncertain. Thereby, uncertainty-related emotions such as fear, anxiety, and −to some extent- sadness lead individuals to perceive a lack of control over the situation or event and to ascribe control to the environment rather than to themselves (Anderson et al., 2019). This emotional uncertainty serves as an internal cue as to whether more processing is required before making a decision (Tiedens & Linton, 2001). Therefore, appraising the situation or event as uncertain increases the tendency to rely on more strategic, analytical, deliberate, and relatively slow information processing strategies. In contrast, appraising the situation or event as certain requires less demand from attentional resources as individuals can easily understand what is happening in the current situation and anticipate how it might end and what might happen next (Smith and Ellsworth, 1985, Tiedens and Linton, 2001). Thus, certainty-related emotions such as disgust, anger, and happiness lead individuals to perceive that they are in control of the situation, ascribing control to themselves rather than the environment. When experiencing certainty-related emotions, individuals tend to use heuristics and follow familiar information without requiring processing additional information to reach a conclusion (Lerner et al., 2015). In sum, based on appraisal theories of emotion, disgust and fear −beyond their idiosyncratic qualities- have a strong potential to promote different information-processing strategies due to their contrasting levels of (un)certainty (Lerner and Keltner, 2000, Smith and Ellsworth, 1985).

Consistent with this claim, previous studies have demonstrated that different specific emotions differentially influence the depth of information processing and decision-making strategies because they have distinct levels of (un)certainty, despite sharing similar valence and arousal levels (Arıkan İyilikci and Amado, 2018, Bagneux et al., 2013, Small and Lerner, 2008, Tiedens and Linton, 2001). For example, compared to uncertainty-associated sadness/fear or control conditions, certainty-related disgust leads individuals to engage in intuitive processing in gambling tasks (Bagneux et al., 2013), to follow fast and intuition-based categorization processes (Arıkan İyilikci, 2018), to use heuristic strategies such as relying on the expertise of a source and stereotypes (Tiedens & Linton, 2001), and to prefer familiar brands compared to novel brands especially when cognitive resources were depleted (Donato, 2021). Thus, specific emotions having different levels of (un)certainty differentially guide decision-making strategies and influence not only the performance at a particular task, but also the depth and content of information processing that leads to that level of performance. Indeed, based on the assumptions of appraisal theories, Levine and Pizarro (2004) argued that information relevant to the appraisal properties of specific emotions appears to be particularly salient in memory and that a complete understanding of the effects of emotion on memory requires taking the appraisal properties of emotions into account. Given that the decision process plays an important role in recognition memory performance (Verde & Rotello, 2007), disgust and fear may differentially influence not only the outcome of memory (i.e., accuracy), but also the underlying decision mechanisms, as their contrasting feelings of (un)certainty guide individuals to different information processing strategies (Smith & Ellsworth, 1985). However, appraisal differences between disgust and fear as a potential explanation beyond their idiosyncratic qualities have been virtually overlooked in most previous memory studies. Therefore, testing specific hypotheses that account for appraisal differences of specific emotions and their potential to drive different information processing strategies would improve our understanding of emotion effects on memory.

Indeed, Kapucu et al. (2024) recently contrasted the recognition memory accuracy and response bias for disgust-, fear-, and anger-related stimuli by focusing on the appraisal properties of these emotions and replicated the disgust-related accuracy and liberal response bias advantage. Nevertheless, this advantage was only observed over anger-related and neutral, but not over fear-related items, indicating that the contribution of (un)certainty differences did not strongly emerge on memory outcome. We note that although Kapucu et al. (2024) accounted for (un)certainty differences among specific emotions and measured memory accuracy independently of response bias, their hypotheses and methods were not directly designed to uncover the decision mechanisms underlying recognition memory. Given that (un)certainty appraisal particularly influences the depth of information processing and underlying decision strategies, to elucidate how appraisal properties contributed to the disgust-related memory advantage, we should focus on the underlying retrieval dynamics, not only on measuring memory outcome.

Although previous studies have addressed the role of encoding and consolidation processes in the disgust-related advantage, they have mostly overlooked the effects of emotion on the retrieval stage of memory. Given that emotions influence decision processes during retrieval (Bowen et al., 2016b, Yüvrük et al., 2023), distinct retrieval dynamics may underlie the disgust-related accuracy advantage.

Indeed, as mentioned above, some previous studies have utilized SDT to better understand the role of emotion in decision styles during the retrieval process (Boğa et al., 2021, Kapucu et al., 2024). Still, they offer only limited insights into the underlying retrieval mechanisms because SDT overlooks the dynamic nature of the decision process. Alternatively, several studies have reported reaction time (RT) data and concluded that emotional information is retrieved more efficiently as participants reach the same level of accuracy with shorter RTs to emotional stimuli (e.g., Jaeger et al., 2017). However, these studies were limited to the valence/arousal dimensions and overlooked the unique properties of specific emotions. Furthermore, conclusions based solely on RT data are rather indirect and do not provide detailed information about the underlying decision mechanisms (White & Kitchen, 2022). Mathematical models have shown that different cognitive mechanisms may underlie the performance even when similar accuracy and RT data are observed and that examining these behavioral data separately ignores individual differences in the speed-accuracy trade-off (Wagenmakers, 2009). Thus, the diffusion model (Ratcliff, 1978) might be the best candidate to uncover the underlying retrieval and decision dynamics of disgust-related advantage in memory by better accounting for accuracy and RT data controlling for individual differences in the speed-accuracy trade-off.

The diffusion model, a sequential sampling model that characterizes simple two-choice decisions as a continuous accumulation of evidence over time, provides a better understanding of the decision mechanisms underlying recognition performance by decomposing it into different components (Ratcliff, 1978, Ratcliff and McKoon, 2008). Namely, the model allows us to estimate the underlying cognitive processes by mapping them to psychologically meaningful parameters. The decision maker's ability to extract information from the stimulus, decision biases, speed-accuracy trade-offs, and non-decision delays, such as the time required for a motor response after a decision is made, can be estimated separately by incorporating the empirical RT distributions along with the proportion of correct and incorrect responses.

The diffusion model posits a dynamic version of the signal detection theory (Ratcliff, 1978). The model defines the decisional component of retrieval as a noisy process in which the evidence accumulates from a starting point (z) toward either the upper (a) or lower (0) response boundaries (see Fig. 1), conventionally representing the ‘Old’ and ‘New’ responses, respectively (e.g., Ratcliff et al., 2004). Evidence is accumulated until it reaches one of the response boundaries, that is until it reaches a sufficient amount to identify a test item as old or new. Once one of the response boundaries is reached, the decision process ends, and a response is executed.

The participants’ a priori response bias before the evidence accumulation begins is represented by the starting point z, which can be defined as its relative position between the upper and lower boundaries. If the starting point, z, is closer to one of the boundaries, then the evidence accumulation is more likely to terminate at that boundary. For a recognition memory task, if z is closer to the upper boundary than the lower one, the participant is biased to respond “Old” more often than “New”, namely endorsing a more liberal response bias. In the diffusion model, the distance between boundaries (i.e., boundary separation a) is characterized as the speed-accuracy trade-off endorsed by participants. The greater the distance, the greater the amount of evidence required to make a recognition decision, indicating a more cautious response style that results in slower but more accurate responses. Conversely, the smaller the distance, the lesser the amount of evidence required to make a recognition decision, indicating a less cautious response style that results in faster but less accurate responses.

The rate of the evidence accumulation supporting either an “Old” or “New” response is defined by the drift rate (v) which represents the quality of information extracted from the stimulus. For a recognition memory task, the memory strength of the test item is mapped by the v (Ratcliff et al., 2004). The greater the memory strength, the higher the absolute value of the v, indicating that the evidence accumulation quickly reaches the response boundary. If the evidence accumulation terminates at the upper boundary, v has a positive value, while if it terminates at the lower boundary, v has a negative value. We note that as the decision process is noisy, within-trial memory strength varies from moment to moment, suggesting that processes with the same mean drift rate may not terminate at the same time (resulting in RT distributions) or at the same response boundary (resulting in a mixture of correct and incorrect responses).

The model allows for separate estimation of drift rates for targets (vtarget) and lures (vlure). Thereby, total drift rates for targets and lures (i.e., vtotal = vtarget + vlure) represent a relative tendency to extract mnemonic information in favor of either an “Old” or a “New” response. vtotal is also an indication of the position of the drift criterion (e.g., Starns et al., 2012), participants’ criterion for how strong memory evidence must be to move evidence accumulation toward “Old” or “New” response boundary. Therefore, it reflects a bias at the level of retrieval processes, rather than at the response level (White and Poldrack, 2014, White et al., 2016). Thus, vtotal allows for the diffusion model to surpass SDT by capturing two sources of biases (memory bias1 vs. response bias) induced by emotion during recognition memory tasks.

When familiarity2 and novelty signals are equally strong during retrieval, vtotal is zero because drift rates for targets are positive while drift rates for lures are negative. Therefore, positive vtotal values indicate enhanced familiarity memory bias that facilitates evidence accumulation toward the “Old” response boundary, while negative vtotal values indicate novelty memory bias that facilitates evidence accumulation toward the “New” response boundary.

In sum, the basic parameters of the diffusion model can account for information processing strategies triggered by specific emotions, as described by appraisal theories. This includes participants' motivational preference for one of the two memory decisions (response bias), captured by starting point z; participants' response style when making a recognition decision (response cautiousness), captured by boundary separation a; and the relative weight given to familiarity vs. novelty signals from memory during retrieval (memory bias), captured by total drift rate vtotal. We note that the model described above was introduced as the basic diffusion model with its four main parameters (v, a, z, to) (Stone, 1960). Yet, current versions of the model estimate other parameters representing across-trial variability in these main parameters (see Ratcliff and McKoon, 2008, Wagenmakers, 2009, for further information about the model).

Only a limited number of emotional memory studies have utilized the diffusion model to investigate the retrieval dynamics of emotional items (Bowen et al., 2016b, Kapucu, 2010, Spaniol et al., 2008). White et al. (2009) showed that the same pattern of accuracy and response time can be observed for emotional and neutral items even though different cognitive mechanisms underlie the processing of emotional stimuli, leading to the misleading conclusion of a null effect of emotion on recognition memory. When accuracy and response time data are analyzed separately, individual differences in the speed-accuracy trade-off may hinder emotion effects. The diffusion model, however, allows us to elucidate what kind of cognitive mechanisms underlie emotional memory by incorporating accuracy and RT data together. For example, the well-documented negativity bias in memory in younger adults (i.e., higher memory accuracy for negative emotional items) tends to decrease with age. Instead, memory accuracy in older adults tends to be higher for positive items than for neutral or negative items (see Mather & Carstensen, 2005). Using the diffusion model, Spaniol et al. (2008) investigated this age-related positive shift in memory by considering two distinct cognitive mechanisms. That is, they distinguished between response bias and memory bias thanks to the diffusion model parameters and questioned which type of bias better contributes to the positivity effect in older adults. Along with traditional accuracy (d') and RT data, Spaniol et al. (2008) estimated the relative position of starting point z between response boundaries and total drift rate vtotal separately for emotional vs. neutral items. Starting point z was interpreted as an indication of motivational preference (response bias) for one of the two memory decisions (i.e., old or new), whereas vtotal was interpreted as a general tendency to extract mnemonic information in favor of either familiarity or novelty signals from memory (memory bias). Although they revealed comparable levels of accuracy and RT for negative, positive, and neutral items, the results of the diffusion model drew a different conclusion. They reported a starting point closer to the “Old” boundary, indicating a general increase in liberal response bias for emotional items compared to neutral ones. Furthermore, the higher vtotal values (i.e., a stronger familiarity memory bias facilitating evidence accumulation toward the “Old” response boundary) was observed for positive items than negative and neutral items in older adults. Thus, the diffusion model supported the differential processing of positive items in memory in older adults despite the null effects of emotion on accuracy and RT data (Spaniol et al., 2008). Furthermore, Bowen et al. (2016b) investigated the negativity effect in memory in younger adults by considering response bias and memory bias by estimating the same parameters of the diffusion model as in Spaniol et al. (2008). They also examined at which stage of the memory process emotion modulation occurs by testing the memory at 1-day and 7-day retention intervals. Namely, they examined the contribution of emotion-specific consolidation to response and memory biases. Bowen et al. (2016b) reported a starting point closer to the “Old” boundary for negatively-valenced items compared to neutral ones, indicating a more pronounced liberal response bias. The more notable results were observed for vtotal values. Negative or highly-arousing items facilitated the information accumulation favoring the “Old” decision (i.e., the higher vtotal) compared to positive or low-arousing items, respectively. Namely, the diffusion model suggested that negative or highly-arousing items enhance a familiarity memory bias independent of actual discriminability. Remarkably, this pattern of vtotal became more pronounced with a longer retention interval, supporting the contribution of emotion-specific consolidation to memory bias, but not to response bias (Bowen et al., 2016b). We note that, despite the valuable contributions of the diffusion model to the emotional memory literature, those studies ignored the effects of specific emotions and their appraisal properties on recognition memory and underlying decision mechanisms, which motivated us to design the present study.

In the present study, our primary aim was to investigate the underlying decision mechanisms of disgust-related memory advantage over fear. Appraisal theories of emotion suggest that this advantage may stem from differences in the unique appraisal properties of these emotions and their role on the decision process (Lerner and Keltner, 2000, Smith and Ellsworth, 1985). As previously mentioned, disgust is a certainty-related emotion while fear is uncertainty-related, despite their similar valence/arousal levels and motivational tendencies. Thus, appraisal theories predict that disgust should increase the reliance on familiarity signals during retrieval and follow heuristics without requiring deliberate and relatively slow information-processing strategies, possibly favoring speed over accuracy, compared to uncertainty-related fear. Although decision-making literature provides striking evidence for differential information processing strategies for disgust and fear due to their contrasting feelings of (un)certainty, most emotional memory studies have ignored the appraisal properties of specific emotions or have not used an appropriate methodological approach to examine the decision process during retrieval. Therefore, the potential influence of (un)certainty differences between disgust and fear on memory decisions has yet to be investigated.

To this aim, we conducted two experiments, each testing similar hypotheses using different research designs. To summarize, the effects of disgust-related, fear-related, and neutral pictures were tested either immediately (Exp 1) or after a 24-hour retention interval that allows for consolidation (Exp 2). We based our hypotheses on the predictions of appraisal theories of emotion, as an alternative approach to the valence/arousal dimensions. Note that we consider appraisal differences as a potential explanation for the observed effects, rather than the sole account. Although disgust and fear are best suited to eliminate potential confounds of irrelevant emotional dimensions, ruling out emotion-specific mechanisms is not possible based solely on a comparison of these two emotions. We estimated the basic parameters of the diffusion model (a, z and vtotal) to disentangle underlying decision mechanisms of the disgust-related memory advantage, along with traditional measures of accuracy and RT. The specific hypotheses were explained in the introduction sections of the corresponding experiments.