PaperlayerView exact revision on arXiv

Source-linked AI summary

Reconstruction and Reflection of Positive Experiences through Resurfacing Laughter-indexed Everyday Moments

Jun Fang, Jiajin Li, Yuntao Wang, Kexin Miao, Susu Wang, Xiaoyu Xie, Kaixin Ji, Yuanchun Shi

arXiv:2609.12642v1cs.HC

TL;DR

Deliberate recording can miss brief or socially absorbing moments, while continuous capture creates extensive records to review. The paper presents LaughAnchor, which indexes laughter with aligned context for later reconstruction and reflection. In a three-week deployment, participants valued captured moments they said they were unlikely to record deliberately.

  • Problem

    Deliberate recording may miss brief or socially absorbing moments, while continuous capture produces large streams that users must later segment and review.

    Source: §1 · p. 2

  • Method

    LaughAnchor uses detected laughter and aligned context to assemble candidate moments during participant-initiated recording, with layered disclosure, user curation, and later resurfacing.

    Source: §1 · p. 3

  • Results

    All 12 participants identified at least one captured moment they would rarely have recorded themselves but later considered worth preserving; participant-equal valued-novel coverage was 43.6%.

    Source: §6.3.1 · p. 20

  • Takeaways & Limitations

    The paper concludes that personal-record value can emerge after capture and need not depend on complete recall.

    Source: §8 · p. 27

    Richer contextual capture raises bystander privacy concerns that layered disclosure and user curation do not resolve during capture and processing.

    Source: §7.3 · p. 27

Abstract

from arXiv · show

Positive everyday moments often escape deliberate recording, while continuous self-tracking can generate extensive records that are difficult to revisit. We explore laughter as a naturally occurring, sparse index for constructing contextualized personal records to support later reconstruction and reflection. A formative study with 12 participants characterized laughter as an affective but semantically incomplete index and informed \textit{LaughAnchor}, a mobile and wearable self-tracking system. During participant-initiated recording, the system assembles detected laughter and aligned context into candidate moments for later reconstruction and reflection, with layered context disclosure, user-controlled curation, and near-term and long-term resurfacing. In a three-week field deployment with 12 participants, passive indexing preserved moments they considered unlikely to record deliberately but valued retrospectively. During resurfacing, participants attributed affective re-experiencing to laughter and used additional context both to reconstruct episodes and to explore already-recalled experiences. Across moments and reviews, resurfacing supported rediscovery and broader awareness of relationships, routines, and emotional states. These findings inform self-tracking designs that use sparse affective indices to organize contextual records for reconstruction and reflection, while keeping interpretation and retention under user control.

1 Introduction

Positive moments can be missed by deliberate recording, while continuous capture creates records that are difficult to review. The paper uses laughter as a sparse but semantically incomplete index, then develops and evaluates LaughAnchor to organize contextualized moments for later reconstruction and reflection.

Motivation

  • Deliberate recording may miss brief or socially absorbing moments, while continuous capture leaves users with large streams to segment and review.The challenge is preserving potentially valuable traces without requiring their value to be known in advance.Source: §1 · pp. 2–3

Laughter as an index

  • Laughter offers a naturally occurring, sparse affective cue, but does not specify what happened, why it mattered, or whether it merits preservation.The paper defines a laughter-indexed moment as a candidate anchored by detected laughter and scaffolded with bounded contextual cues.Source: §1 · p. 3

Research gap

  • Rather than representing laughter alone, the approach uses it to index a surrounding episode that can be grounded through context and user interpretation.Prior work explored tangible laughter representations and visualizations of its temporal, spatial, and social aspects.Source: §1 · p. 3

Approach

  • A formative study with 12 participants informed LaughAnchor, which assembles detected laughter and aligned context during participant-initiated recording for later review.The system supports layered disclosure, user curation, and near-term and long-term resurfacing.Source: §1 · p. 3

Contributions

  • In a three-week deployment with 12 participants, LaughAnchor complemented existing recording practices and supported affective reconnection, reconstruction, exploration beyond recall, and reflection across moments.These are the empirical contributions described by the paper.Source: §1 · p. 3

2 Background and Related Work

Prior self-tracking approaches trade in-the-moment effort against the volume and interpretive demands of captured records. This work positions laughter as a sparse affective index, then connects resurfacing to user-led reconstruction and reflection.

2.1 Self-Tracking for Positive Everyday Moments

  • Positive everyday moments may be brief and overlooked, yet resurfacing them can support affective re-experiencing, reflection, and appreciation of routine life.The cited work frames this as a motivation for self-tracking positive experiences.Source: §2.1 · p. 4
  • Deliberate recording can miss brief or socially absorbing moments, whereas passive approaches reduce capture effort but shift selection and interpretation to later review.LaughAnchor explores an intermediate approach: laughter indexes contextualized candidate moments for later reconstruction and user interpretation.Source: §2.1 · p. 4

2.2 Laughter as an Affective yet Ambiguous Index

  • Laughter conveys affective and social cues, but alone does not establish positivity, personal importance, or reflective value.The paper therefore treats laughter as a candidate anchor for potentially positive moments.Source: §2.2 · pp. 4–5
  • Earlier laughter systems represented or visualized laughter; this work shifts attention to contextual reconstruction, user-led interpretation, and later resurfacing.LaughAnchor implements layered context disclosure, user-controlled curation, and near-term and long-term resurfacing.Source: §2.2 · p. 5

2.3 Resurfacing for Reconstruction and Reflection

  • Reconstruction rebuilds an episode from cues, while reflection can develop meaning even when recall is incomplete; resurfacing offers an opportunity for these processes, not a guarantee.Their occurrence depends on timing, context, and user control.Source: §2.3 · p. 5

3 Apparatus and Laughter-Indexing Pipeline

The system uses a glasses-mounted camera–microphone module and smartphone to capture aligned first-person audiovisual context. During participant-initiated sessions, it detects laughter and organizes it with bounded context into candidate moments for review.

Apparatus

  • A glasses-mounted camera–microphone module captures head-aligned first-person audiovisual context and connects to an Android smartphone.The phone supplies power, stores recordings, and runs the companion application.Source: §3 · p. 6, Figure 2 · p. 6

Laughter-indexing pipeline

  • During participant-initiated recording, the application streams audio for timestamped laughter detection and retains events above a study-specific confidence threshold.Temporally adjacent bouts are grouped and aligned with the locally stored recording.Source: §3 · p. 6
  • Each candidate associates detected laughter with bounded audiovisual and spatiotemporal context for subsequent review.The shared sensing architecture and indexing logic were adapted to each study's parameter settings.Source: §3 · p. 6

4 Formative Study

The formative study examined how laughter and aligned context support reconstruction, and how people judge the meaning and resurfacing of captured moments. Participants described laughter as affectively evocative but insufficient by itself; context, temporal distance, and user interpretation shaped reconstruction and reflection.

4.1–4.2 Study design and procedures

  • The formative study used participants' own laughter episodes and progressive context disclosure to examine reconstruction and reflection without assigning meaning in advance.Participants completed interviews, recorded self-selected activities for seven days, and reviewed near-term and long-term moments; the review was not an assessment of objective memory accuracy.Source: §4 · p. 6, §4.2 · pp. 7–8

4.3 Data analysis

  • The team used thematic analysis and scored reconstructions across five internal-detail categories on a 0–15 total scale.Two authors independently coded 60% of the qualitative data before the team coded the remainder; category scores covered event, place, time, perceptual information, and emotion.Source: §4.3 · pp. 8–9

4.4.1 F1

  • Laughter reinstated an episode's emotional tone but did not reliably identify the specific event, especially for distant episodes or repetitive activities.Reconstruction emerged when laughter aligned with episode boundaries and proximal context; nearby speech and audiovisual cues helped recover interactions and triggers.Source: §4.4 · pp. 9–10

4.4.2 F2

  • Temporal distance changed reconstruction needs: long-term reviews often began with affective re-entry, while near-term reviews more often required disambiguating similar interactions.Long-term laughter gains were largest for emotion at L0 and event and perceptual detail at L3; near-term gains were more broadly distributed, with diminishing gains after a coherent scene formed.Source: §4.4 · p. 10, Figure 4 · p. 10

4.4.3 F3

  • Passive capture reduced in-the-moment effort and preserved brief, ordinary moments participants might not have recorded but later valued.It also accumulated ordinary or repetitive moments, broadening the candidates available for later consideration.Source: §4.4 · pp. 10–11

4.4 Findings

  • Laughter's occurrence did not determine a moment's reflective value or whether it should be retained; participants wanted to interpret and supplement records later.Judgments depended on context and personal significance, and participants cautioned that generated summaries could replace rather than support their own meaning-making.Source: §4.4 · p. 11

4.4.4 F4

  • Near-term resurfacing reinforced or clarified ongoing positive feeling, while long-term resurfacing helped participants rediscover ordinary moments and consolidate meaning.Participants described these benefits as complementary, without reporting fundamental changes in their overall evaluation of life.Source: §4.4.4 · p. 11

4.4.5 F5

  • Participants preferred resurfacing to be optional, low-interruption, and easy to ignore, while allowing occasional prompts to surface moments they would not deliberately retrieve.Situated cues, such as returning to a location, were one suggested way to fit resurfacing into everyday activity.Source: §4.4.5 · p. 11

5 LaughAnchor System Design

LaughAnchor turns detected laughter and surrounding context into candidate Moment Cards, while leaving interpretation and retention to users. Its design pairs layered context disclosure with near- and long-term resurfacing, optional access, and user-controlled curation.

5.1 Key Elements of LaughAnchor

  • Context disclosure progresses from laughter-based recognition to episode reconstruction, with the initially visible cues adapted to near- or long-term review.Long-term reviews default to laughter audio, time, and location; near-term reviews bundle these with a photo or video preview and surrounding audio.Source: §5 · p. 12, §5.1.1 · p. 13
  • Users retain authority over a candidate’s meaning and preservation, including the ability to revise retention and resurfacing choices or add personal accounts later.Candidates are retained by default unless deleted; users can save them with or without proactive resurfacing and edit summaries or comments asynchronously.Source: §5 · p. 12, §5.1.1 · p. 14
  • The design supports complementary near- and long-term resurfacing while keeping prompts optional, dismissible, and sensitive to the current context.A retained moment can return at different temporal distances; users may also receive a suggestion when they revisit its capture location.Source: §5 · pp. 12–13, §5.1.1 · p. 14
  • During participant-initiated recording, the system groups detected laughter with bounded contextual cues into candidate Moment Cards for later review.Users can supplement or delete candidates and decide whether retained moments may receive future resurfacing suggestions.Source: Figure 5 · p. 13, §5.1.1 · p. 14
  • Calendar, Timeline, and Map views support self-initiated access, while laughter frequency and duration summaries offer a cautious entry point for observing routines and social rhythms.The interface leaves the personal significance of these quantities for users to interpret rather than treating more laughter as a better outcome.Source: §5.1.1 · p. 14

5.2 Apparatus and Implementation

  • The field prototype retained the shared sensing and laughter-indexing architecture while primarily revising the companion app’s interface and display logic.System parameters were also adjusted using observed performance and participant feedback; benchmarking, context acquisition, and study-specific settings are reported in Appendix A.Source: §5.2 · p. 14

6 Field Study

The three-week in-the-wild deployment examined how laughter-indexed resurfacing surfaced participant-valued moments they considered unlikely to record deliberately and supported affective reconnection, episode reconstruction, and exploration of additional context. It also examined reflection on broader patterns in relationships, routines, and emotional states, alongside exploratory well-being findings and participants’ descriptions of positive attention, convenient capture, and favorable usability.

6.2 Data Collection

  • Evaluation paired participants’ habitual practices with post-resurfacing questionnaires, usage logs, and interviews rather than imposing a standardized journaling comparator.The measures covered individual reviews, context use, and well-being; interview findings were interpreted alongside questionnaire responses and usage patterns.Source: §6.1 · p. 15, §6.2.4 · pp. 18–19

6.1 Participants and Study Design

  • The study enrolled 12 participants for a flexible three-week deployment, with no fixed daily schedule or minimum session duration and recommended minimums for recording and both review types.Participants chose when and where to record and review moments, and could continue, suspend, or modify their existing practices.Source: §6.1.1 · p. 16, §6.1.2 · p. 16

6.3 Results

  • 24/54 near-term reviews (44.4%), spanning 23 distinct moments, met valued-novel criteria; participant-equal valued-novel coverage was 43.6% (95% CI [25.0%, 63.1%]).Nine of 12 participants encountered at least one valued-novel moment.Source: §6.3.1 · p. 20, Table 6 · p. 21
  • Passive indexing captured ordinary or interaction-absorbed moments before recording intent, helping lower-frequency recorders preserve whole experiences and higher-frequency recorders capture undocumented interactions between selected landmarks.Participants described moments they would rarely have recorded themselves but later valued, including experiences missed through barriers to deliberate capture and conversations between planned photos or posts.Source: §6.3.1 · pp. 20–21
  • 6.15 out of 7 was the participant-equal Affective Re-experiencing mean across 95 reviews, versus a 5.53 habitual-practice rating (+0.62 points).Laughter Entry averaged 6.06 out of 7 and Added Information 5.72; in nine near-term reviews with external records, laughter's stronger-emotion rating averaged 6.67 out of 7. Laughter playback occurred in 85 of 95 questionnaire-linked reviews.Source: §6.3.2 · p. 22
  • Participants described laughter as quickly restoring an episode's emotional state and vitality, while nearby speech helped situate the response in the interaction.Participants attributed atmosphere, emotional intensity, laughter triggers, and interactional information to laughter; they described it as a central affective cue rather than an interchangeable contextual one.Source: §6.3.2 · p. 22
  • 6.00 out of 7 for Identification, 5.77 for Reconstruction Detail, and 6.02 for Cue Sufficiency indicate participants generally identified and reconstructed episodes through resurfaced context.In 65 of 95 reviews (68.4%), participants did not report needing additional context for reconstruction; 44 of those reviews involved further exploration after recall.Source: §6.3.3 · p. 22
  • Long-term resurfacing was associated with a +0.81-point difference in Rediscovery Awareness versus habitual practice (95% CI [0.38, 1.27], p = .006, rrb = .945).System ratings were descriptively similar to habitual practice for Identification (6.00 vs. 5.75) and Reconstruction Detail (5.77 vs. 5.61), with no reliable paired differences; added context was more frequently needed for long-term than near-term reconstruction.Source: §6.3.3 · p. 23
  • Participants used laughter as an entry point to recover triggers, conversation, and unfolding episodes, while long-term reviews more often required added context than near-term reviews.Near-term resurfacing could be checked against accessible memories; long-term resurfacing helped recover details that participants said they had nearly forgotten.Source: §6.3.3 · p. 23
  • Reviewing multiple moments helped participants recognize recurring people, activities, and emotional states and develop appreciation for relationships and ordinary family conversations.Participants described awareness as noticing patterns across moments and meaning as interpreting their personal significance.Source: §6.3.3 · p. 23
  • WHO-5 scores increased from 56.7 at baseline to 74.3 mid-deployment and 84.3 at completion, a significant baseline-to-completion increase (Δ = +27.7, 95% CI [12.33, 42.67], p = .011).Neither review rhythm nor review depth was significantly associated with baseline-to-completion change, so interviews examined participants’ interpretations and attributions.Source: §6.3.4 · p. 24
  • Participants described LaughAnchor as making small positive experiences more salient and reinforcing immediate positive feelings, while also helping them notice well-being fluctuations alongside everyday circumstances.Some associated perceived mood improvements with continued system use; P12 described positive reinforcement despite increased work pressure, alongside broader well-being changes linked to that pressure.Source: §6.3.4 · p. 24
  • LaughAnchor received a favorable SUS score (M = 83.5, SD = 8.4); participants accepted 98 of 111 resurfacing notifications (88.3%) across the deployment.Across 199 identifiable user-initiated review episodes, access was concentrated in the evening but varied by participant. Participants valued passive capture for its convenience and expressed interest in continued use.Source: §6.3.5 · pp. 24–25, Figure 11 · p. 25

7 Discussion

Laughter can invite affective engagement even before precise recall, while accessible context supports reconstruction and further exploration. Accumulated moments also support awareness of everyday emotional experience, within limits shaped by study duration, participant range, and capture privacy.

7.1 Affective Value and Contextual Exploration Beyond Precise Reconstruction

  • Laughter could offer enjoyment during resurfacing even when participants could not recall the episode precisely.One participant laughed again while listening to a recording without knowing what the joke was; laughter-indexed clips also prompted curiosity and unexpected rediscovery.Source: §7.1 · p. 25
  • Participants used additional context for reconstruction in 30 of 95 reviews and for further exploration after recall in 44 reviews.More expansion did not necessarily indicate inadequate initial cues, and successful recall did not make remaining material redundant.Source: §7.1 · p. 25
  • Contextual materials should remain available for reconstruction and continued exploration, while users decide when their purposes are met.A sparse affective trace can serve as an entry point without requiring complete recall or making recall the endpoint.Source: §7.1 · pp. 25–26

7.2 From Captured Positive Moments to Everyday Awareness in Self-Tracking

  • Accumulated laughter-indexed moments gave participants material for noticing emotional changes, which they attributed to everyday circumstances rather than recording.The records provide a selective view whose significance depends on how users relate it to lived circumstances.Source: §7.2 · p. 26
  • Participants reported seeking enjoyable experiences and approaching activities more positively while recording, but these accounts do not establish a causal wellbeing effect.Recording and resurfacing may scaffold habits of noticing and reflection, though persistence beyond system use remains a future question.Source: §7.2 · p. 26

7.3 Limitations and Future Work

  • The study’s primarily student and working-adult participants and three-week deployment limit insight into other contexts and longer-term reflection.Longer studies could examine how reconstruction changes as memories become more distant and how the system relates to habitual recording and review.Source: §7.3 · p. 26
  • Different event selection and measurement contexts limited direct comparison between habitual-practice and post-resurfacing assessments.The prototype’s external glasses-mounted camera–microphone and wired phone connection also added setup and handling demands.Source: §7.3 · p. 26
  • Layered disclosure and curation support control over records but do not resolve bystander privacy concerns during passive capture and processing.Richer context can aid reconstruction while also collecting additional personal, social, and interactional information.Source: §7.3 · p. 27

8 Conclusion

LaughAnchor uses laughter as a sparse affective index to build contextualized personal records for later reconstruction and reflection. In a three-week deployment, laughter provided an affective entry point while context supported reconstruction and exploration, and resurfacing prompted enjoyment, curiosity, and reconsideration of ordinary experiences.

8 Conclusion

  • LaughAnchor uses laughter as a sparse affective index to construct contextualized personal records for reconstruction and reflection.The mobile and wearable system connected laughter-indexed capture with contextual grounding, user curation, and later resurfacing.Source: §8 · p. 27
  • Laughter offered an affective entry point, while contextual cues supported episode reconstruction and continued exploration after recall.The value of a personal record can emerge after capture and need not depend on complete recall.Source: §8 · p. 27
  • Resurfacing can offer enjoyment, spark curiosity, and prompt reconsideration of ordinary experiences while leaving users to decide their significance.Designing for these possibilities involves letting people choose how much to reconstruct and when to explore further.Source: §8 · p. 27

A.1 Laughter Detector Selection

The detector comparison favored reliable timestamp boundaries and high precision for assembling laughter moments with aligned context. Open-source models performed poorly on timestamp-level detection, and Speechmatics’ precision–recall trade-off matched participants’ greater tolerance for missed laughter than false detections.

A.1 Laughter Detector Selection

  • The benchmark compared four open-source laughter detectors and an API to inform the backend choice.The comparison covered event-level and timestamp-level detection, with event-level performance also considered with and without ASR filtering.Source: §A.1 · pp. 30–31
  • All four open-source methods had timestamp-level F1 ≤0.473, despite reasonable event-level precision.Timestamp boundaries mattered because LaughAnchor aligns each detected event with surrounding context when assembling a moment card.Source: §A.1 · p. 31, Table 10 · p. 32
  • Speechmatics was selected for relatively low recall but consistently high precision, reducing false positives that could confuse users and undermine trust.Participants were more tolerant of missed laughter than false detections, consistent with using laughter as an opportunistic rather than exhaustive index.Source: §A.1 · pp. 31–32

A.2 Event Construction and Post-Processing

The appendix describes modality-specific rules for constructing formative-study materials and reports field-study adjustments to candidate-moment grouping and detection confidence.

A.2 Event Construction and Post-Processing

  • The formative study used the Section 3 apparatus and capture pipeline, with modality-specific rules to generate its materials.Table 11 summarizes those rules.Source: §A.2 · p. 32

A.3 Context Capture Configuration for the Formative Study and Field Study

  • In the field study, EVENT_CLUSTER_INTERVAL rose from 600 to 1200 s, setting the maximum gap for grouping consecutive laughter bouts into one candidate moment; LAUGHTER_CONFIDENCE_THRESHOLD rose from 0.60 to 0.70.The interval was increased to reduce fragmentation, since the same ongoing activity was sometimes split across candidate records; it is a temporal grouping heuristic, not a detector of semantic episode boundaries.Source: §A.3 · p. 33

A.4 Resurfacing Scheduling Policy

LaughAnchor resurfaces saved moments through daily and location-based channels, with prioritization and frequency limits intended to reduce redundant notifications.

A.4 Resurfacing Scheduling Policy

  • At 19:30 local time, daily resurfacing selects one eligible moment from the previous day and one from seven days earlier, only if the user enabled saving for suggestions.The two selections are handled independently as near-term and long-term resurfacing.Source: §A.4 · p. 33
  • Location resurfacing triggers when users re-enter a 50-m boundary around a stable place cluster for an eligible historical moment.Only events older than six hours are considered, with at least two hours between location notifications.Source: §A.4 · p. 33
  • When several moments qualify, the system first prioritizes user-added content and then favors candidates with more media within the same priority level.User-added content includes notes, recordings, photos, or social information; media includes automatically captured and user-contributed audio, photos, and videos.Source: §A.4 · p. 33
  • Daily and location channels have independent on/off controls, enabled by default, and deduplicate alerts to at most one per category or place cluster per day.Daily categories are near-term and long-term.Source: §A.4 · p. 33

B Study Instruments

The study instruments covered participants’ habitual recording and review practices, then assessed how records and laughter-indexed moments supported identification, reconstruction, affective re-experiencing, and broader awareness.

B.1 Questionnaire for Field Study

  • The pre-study questionnaire asked about recording frequency and methods, unrecorded moments and reasons, deliberate review, discontinued tools, and perceived effort or interruption.Participants also rated whether valuable moments were missed afterward and whether they tended to record only experiences that already seemed important.Source: §B.1.1 · pp. 33–34
  • Participants revisited representative recent and one-to-two-week-old records using their usual methods, then rated identification, reconstruction detail, affective re-experiencing, and rediscovery awareness.Multi-item measures with Cronbach’s alpha of at least 0.70 were combined into scores.Source: §B.1.1 · p. 34
  • Post-resurfacing ratings assessed whether laughter-indexed moments were identifiable, reconstructable, emotionally re-experienced, and sufficiently supported by the presented cues.Items also asked whether laughter itself was an entry point or added information beyond photographs or text.Source: §B.1.2 · pp. 34–35
  • The post-resurfacing instrument asked what experiential qualities laughter added, including atmosphere, emotional intensity, the laughter trigger, and interaction.Participants also reported whether they accessed additional context to complete reconstruction or to keep exploring after recall.Source: §B.1.2 · p. 35
  • Long-term ratings measured rediscovery awareness, while near-term questions covered prior external records, likelihood of deliberate recording, and whether laughter re-evoked emotion more strongly than an existing record.The questionnaire also asked about the personal value of the moment.Source: §B.1.2 · p. 35

B.2 Post-Study Interview Questions

The approximately 20-minute semi-structured post-study interviews asked about capture and coverage, laughter’s contribution, reconstruction and reflection, and everyday-life awareness.

B.2 Post-Study Interview Questions

  • Interviewers reviewed each participant’s prior recording practices and event-level questionnaire responses, and tailored only the first interview part to prior recording practices.All participants completed Parts 2–4.Source: §B.2 · p. 35
  • The capture-and-coverage questions explored why participants recorded little or whether laughter-indexed capture preserved moments or qualities their usual practices would miss.Prompts asked whether moments were worth retaining and why, and whether participants would normally have recorded them.Source: §B.2 · p. 36
  • Questions about laughter asked how laughter or surrounding voices affected re-experiencing emotion, atmosphere, or interaction, including compared with other media and across review timing.Participants were also asked what would differ if laughter were removed while other captured content remained.Source: §B.2 · p. 36
  • Prompts asked whether reviews led beyond recall to new understanding of an event, relationship, or recent life, and whether reconstruction alone could still be valuable.Participants compared near-term and long-term reviews and whether review purpose or value changed over time.Source: §B.2 · p. 36
  • The final interview part asked whether system use affected views of everyday life, positive experiences, relationships, emotional states, mood, attention, or wellbeing.Participants were asked to attribute any change to recording and reviewing, experiences during the study, both, or another factor.Source: §B.2 · p. 36
Loading 2609.12642v1…