Humans spend the majority of their lives sleeping, daydreaming, mind wandering, or otherwise not paying attention to the external world (Killingsworth and Gilbert, 2010, Wamsley et al., 2023). These “offline” states may have multiple functions, including the consolidation of recently encoded memory. It is well established that sleeping after learning improves memory (Diekelmann and Born, 2010, Mednick, Makovski, Cai, & Jiang, 2009; Payne et al., 2008, Plihal and Born, 1997, Stickgold and Walker, 2013, Tucker et al., 2006). However, recent research shows that a period of eyes closed rest can also retroactively facilitate memory, in a manner similar to sleep (Dewar et al., 2012, Gottselig et al., 2004, Wamsley, 2019, Wang et al., 2021). This effect can be induced by just a few minutes of rest, apply to a wide variety of forms of learning and memory, and last for at least 2 weeks (Brokaw et al., 2016, Craig et al., 2015, Craig et al., 2016, Dewar et al., 2012, Wamsley and Summer, 2020). In the current study we asked whether, in addition to strengthening veridical memory, eyes-closed waking rest might also affect the development of false memory.

Rest quantitatively strengthens memory in a wide range of tasks. Most commonly, this has been observed for verbal declarative memory tasks, in which memory is improved more following rest compared to an active wake condition (Brokaw et al., 2016, Dewar et al., 2012, Dewar et al., 2014, Martini et al., 2018, Wamsley et al., 2023, Wang et al., 2021). However, rest has also been reported to strengthen spatial memory (Craig et al., 2015, Craig et al., 2016), procedural memory (Humiston and Wamsley, 2019, Wang et al., 2021), and emotional memory (Hudachek & Wamsley, 2024). In a handful of studies, these quantitative benefits of rest have been behaviorally indistinguishable from those of sleep (Gottselig et al., 2004, Wang et al., 2021). However, it remains unclear whether post-training rest can replicate all the same behavioral memory effects that have been attributed to sleep.

Although there is strong evidence that sleep qualitatively transforms memory over time, we have a limited understanding whether rest also qualitatively transforms memory. Memory transformation, defined as the qualitative change (rather than the quantitative strengthening) of memory over time, is exemplified by a number of different phenomena in which the features of memory are altered, rather than merely strengthened (Ellenbogen et al., 2007, Payne et al., 2008, Stickgold and Walker, 2013, Wagner et al., 2004). This includes the consolidation of emotional components of scenes (Denis et al., 2022, Payne et al., 2008), the development of insight (Debarnot et al., 2017, Wagner et al., 2004, Yordanova et al., 2012), the boosting of creativity (Horowitz, Esfahany, Gálvez, Maes, & Stickgold, 2023) and memory for information that participants know they will need in the future (Scullin and McDaniel, 2010, van Dongen et al., 2012, Wilhelm et al., 2011), all of which are enhanced by sleep.

Another key type of memory transformation is the development of false memory over time. Sleep has been reported to influence the formation of false memories (Diekelmann et al., 2010, Fenn et al., 2009, Newbury and Monaghan, 2019, Payne et al., 2009). Notably, there is conflicting evidence on whether sleep increases (Payne et al., 2009, McKeon et al., 2012, Shaw and Monaghan, 2017, Mak et al., 2023) or decreases false memory formation (Diekelmann et al., 2010, Fenn et al., 2009, Lo et al., 2014). One proposal for how sleep could increase false memory is its role in “gist extraction”, a process by which the brain extracts the central features of an experience at the expense of accurately remembering details (Ellenbogen et al., 2007, Hu et al., 2006, Michael Lampinen et al., 2006; Payne et al., 2008, Payne et al., 2009). This idea is especially relevant to the Deese-Roediger-McDermott (DRM) paradigm false memory task, where participants falsely remember seeing a “critical lure” that describes the gist of a list of semantically related words. Alternatively, conflicting observations that sleep decreases false memory could be explained by enhanced consolidation of veridical, item-specific details during sleep, which in turn would enhance source monitoring at retrieval (Fenn et al., 2009).

One factor that may explain these divergent results is whether memory is tested using free recall or recognition (Newbury & Monaghan, 2019). One of the most common methods used to study false memory in the laboratory is the DRM paradigm (Deese, 1959, Roediger and McDermott, 1995), in which participants learn lists of semantically related words, and afterwards falsely recall strong semantic associates of those lists as having been among the studied words. A meta-analysis of the effect of sleep on false memory in the DRM paradigm showed that studies using a recall test saw false memory formation increase following sleep (Newbury & Monaghan, 2019). In contrast, there was no significant meta-analytic effect of sleep for recognition tests. Still, there are some individual studies showing reduced false memory production after sleep, most of which tested via recognition rather than recall (Diekelmann et al., 2010, Fenn et al., 2009, Lo et al., 2014).

It is unclear why sleep might have differential effects on false memory when tested by recall vs. recognition, but one possibility is that recall is more cognitively demanding, which could cause it to benefit more from a period of sleep (Kuriyama et al., 2004, Schmidt et al., 2006). Another possibility relates to the difference in retrieval routes available. In a recognition test, participants may rely on both recollection and familiarity when prompted with a word. In contrast, in a recall test participants can only rely on recollection. Potentially, sleep-related offline consolidation could affect recall and recognition differently. Meanwhile, Newbury and Monaghan propose that divergent results for recall vs. recognition are caused by increased source monitoring during recognition, relative to recall testing (Newbury & Monaghan, 2019). Because participants are prompted with words in the recognition test, they may be more likely to engage in critical thought about the source of the word. Potentially, sleep could affect recall and recognition differently by having differential effects on source monitoring.

Like sleep, rest may benefit memory by facilitating the neural-level reactivation of memory networks (Tambini and D’Esposito, 2020, Tambini and Davachi, 2013, Tambini and Davachi, 2019, van de Ven et al., 2016, Wamsley, 2022). It is now well-documented in both animal and human studies that patterns of neural activity first observed during the active encoding of new information are later replayed offline, during both sleep and rest (Carr et al., 2011, Foster and Wilson, 2006, Ji and Wilson, 2007, Lee and Wilson, 2002, Schreiner and Staudigl, 2020, Staresina et al., 2013, Tambini and Davachi, 2019). Furthermore, in both rest and sleep, neural-level reactivation of recent memory traces predicts subsequent performance (Bang et al., 2018, Peigneux et al., 2004, Schreiner et al., 2021, Staresina et al., 2013). Thus, the offline reactivation of memory is a plausible mechanism not only for sleep’s beneficial effect on memory, but also for how resting wakefulness might benefit memory.

Of course, there are also major neurophysiological and psychological differences between rest and sleep. On the neurophysiological level, one example is the presence of sleep spindles, which are associated with increased consolidation after sleep (Mednick et al., 2013, Schabus et al., 2004) but are by definition absent during wakefulness. On the psychological level, subjective experience during waking rest is less hallucinatory and more likely to be focused on current environmental stimuli (Brokaw et al., 2016, Wamsley, 2022). While rest and sleep have been found to equivalently benefit memory in some circumstances (Gottselig et al., 2004, Wang et al., 2021), these neurophysiological and psychological differences could lead these two states to play differential roles in the memory consolidation process.

If rest and sleep indeed affect memory via the same neural mechanism, rest should lead to the same qualitative transformations of memory over time that have been reported for sleep. In support of this hypothesis, a small handful of studies have reported qualitative transformation of memory following rest (Craig et al., 2016, Craig et al., 2018, Hudachek and Wamsley, 2024). Craig et al. reported that rest promotes the development of insight into a hidden rule in the number reduction task (NRT) used by Wagner et al. (2004) to demonstrate the effect of sleep on insight. In this study, participants who rested with their eyes closed after being trained on the NRT were more likely to figure out a hidden shortcut allowing them to quickly solve numerical problems, compared to those who completed a distractor task after training (Craig et al., 2018). Rest has also been reported to facilitate the discovery of shortcuts in a spatial navigation task (Craig et al., 2016). Mirroring the selective effect of sleep on emotional memory, in participants with high levels of trait anxiety, a period of waking rest preferentially improved memory for emotional information (Hudachek & Wamsley, 2024). But studies examining the qualitative transformation of memory following rest remain limited, and the effect of wakeful rest on the formation of false memories has yet to be examined.

In the present study, we asked whether a short period of eyes-closed quiet rest following learning would affect the formation of false memories in the DRM paradigm. Following prior research, we hypothesized that resting after learning would benefit veridical memory, corroborating previous findings (Brokaw et al., 2016, Dewar et al., 2012, Payne et al., 2009, Wamsley, 2022). Second, based on the prior observations discussed above (Mak et al., 2023, Payne et al., 2009, Shaw and Monaghan, 2017) we hypothesized that resting after learning would increase the formation of false memories when using a recall test but decrease false memories when using a recognition test.