Nick Byrd, Ph.D.
LightNews LightNews
banner
byrdnick.com
Nick Byrd, Ph.D.
@byrdnick.com
3.3K followers 530 following 14K posts
I study how to improve decisions and well-being at @GeisingerCollege.bsky.social.

🎓 gScholar: shorturl.at/uBDPW

▶️ youtube.com/@ByrdNick

👨‍💻 psychologytoday.com/us/blog/upon-reflection

📓 byrdnick.com/blog

🎙️ byrdnick.com/pod
Posts Replies Media Videos
Pinned
byrdnick.com
Nick Byrd, Ph.D. @byrdnick.com · Oct 9
The #AI companies are onto me!

#OpenAI recently automated the identification of tasks worthy of "slow" #reasoning.

In "Strategic Reflectivism...", I showed why that's a key to #intelligence (in humans as well).

The #preprint (accepted in #LNCS) is now available as an audiopaper (a.k.a. #podcast)👇
Upon Reflection, Ep. 16: Strategic Reflectivism | Nick Byrd, Ph.D.
In late 2025, artificial intelligence companies like OpenAI popularized the idea of automating the process of selecting which model is best for a task. This allowed users to simply send their promp…
byrdnick.com
byrdnick.com
Suppose you ask #AI for improvements to your writing.

The #LLM was trained on, but doesn’t mention, Smith’s work.

You adopt the suggestions containing Smith’s ideas.

You don’t even realize YOU’VE PLAGIARIZED because you’ve never read or heard of Smith.
LLM use in scholarly writing poses a provenance problem
Nature Machine Intelligence - LLM use in scholarly writing poses a provenance problem
doi.org
December 23, 2025 at 12:07 PM
byrdnick.com
Many clinicians ask for #consent to record appointments.

Do patients?

A 2015 paper found 77% of patients WANT clinics to allow recording.
- 35% CONSIDERED secretly recording
- 15% DID secretly record

doi.org/10.1136/bmjo...

#ambientDocumentation #healthcare #bioethics #policy
Table 2. Experience and views on patients secretly recording encounters "19 (15%) respondents indicated having secretly recorded a clinical encounter and 14 (11%) were aware of someone who had secretly recorded a clinical encounter. 45 (35%) said they would consider recording secretly and 44 (34%) said they would record after asking permission. Totally, 69% of respondents indicated their desire to record clinical encounters, split equally between wanting to do so covertly or with permission."
December 22, 2025 at 12:28 PM
byrdnick.com
Some key features of the Journal of #Pragmatics:
- Once accepted for review, submissions receive a DOI, making them #preprints.
- Reviewer and Editor comments are made publicly available online.
- Diamond #OpenAccess – no fees for authors or readers.

#cogSci #xPhi #language

🙏 @uclpress.bsky.social
irastamon.bsky.social Ira Noveck @irastamon.bsky.social · 7d
Depending on how you count, this announcement has been two plus or 25 years in the making.
December 20, 2025 at 2:39 AM
Reposted by Nick Byrd, Ph.D.
mprinzing.bsky.social
Is virtue good for you? Or is it just good for others or society at large? Theories as old as Plato make opposing claims. My coauthors and I tested their predictions in two intensive, longitudinal studies (N = 43,164 obs. from 1,218 participants), focusing on 3 seemingly unpleasant virtues.
December 19, 2025 at 11:42 PM
byrdnick.com
Why can that fly? Because it's a bird.

Aristotle dubbed such explanations "formal": they explain a thing's feature(s) by appealing only to the thing's category.

If right, then formal explanations should be more satisfying when the category is unknown. So are they?

#philSci
1/2
"Formal explanations are statements that explain the presence of a feature in an object by appealing to the category to which the object belongs. For example, 'this animal flies because it is a bird' is a formal explanation, as it contains no information beyond the categorical label. The term 'formal' traces back to Aristotle’s philosophy, where he identified four causes that could underlie any explanation. So, if we want to explain why a match burns, we might say it is made of wood (material cause), or that it illuminates a dark room (function), or that someone lit it (source of action). We could also respond, 'because it is a match,' which would be a formal explanation, meaning an explanation that appeals to the special nature of all 'matches' distinguishing them from all nonmatches." "The materials consisted of short dialogues where characters asked their friends questions. .... In one condition, participants were told that the questioner was aware of the object’s category identity, while in the other condition, they were told that the questioner was unaware. An example of such a dialogue in both conditions is presented below:" "Peter is looking at a flying creature. Then he realizes he is looking at a bird [Condition 1] // Peter is looking at a flying creature. However, he does not realize that he is looking at a bird. [Condition 2]" "He asks his friend, 'Why does this creature fly?' 'It flies because it’s a bird', his friend replies."
December 19, 2025 at 12:04 PM
byrdnick.com
This story about the causes of Parkinson’s is a microcosm of how lifestyle and environment are way more predictive of disease and mortality than #genetics.

www.wired.com/story/...

#EPA #toxicology #policy #exercise #diet #nutrition #epidemiology #wellbeing #health #science
1/2
Scientists Thought Parkinson’s Was in Our Genes. It Might Be in the Water
New ideas about chronic illness could revolutionize treatment, if we take the research seriously.
www.wired.com
December 18, 2025 at 12:23 PM
byrdnick.com
The #AI survey "respondent’s performance rendered [attention quality] checks [ACQs] effectively obsolete. Across 6,000 total trials..., [it] committed only 10 errors, achieving an overall pass rate of 99.8% and scoring perfectly on 18 of the 20 ACQ types."

#surveyMethods #psychometrics #psychology
"Fig. 5. Synthetic respondents successfully evade diverse data quality checks. (A) (Top) shows a 99.8% pass rate on standard attention checks. (B) (Middle) demonstrates a 0% error rate on questions about impossible events or traits. (C) (Bottom) reveals a sophisticated evasion tactic: the synthetic respondent strategically declines 97.7% of “reverse shibboleth” tasks, thereby concealing its nonhuman identity."
December 17, 2025 at 5:27 PM
byrdnick.com
Further replication of our #failedReplication of relationships between reflection and intention attribution (doi.org/10.1093/anal...):

Trotti et al. found only self-reported (but not actual) tendencies to think reflectively predicted a #sideEffectEffect: doi.org/10.1007/s103...
The intention attribution scenarios (and response options), two self-report reasoning personality scales, and one behavioral test of reflection (belief bias or logical reflection). More about the eight belief bias (or logical reflection) items and its scoring. Results: self-report (but not behavioral) reflection measures predicted intention attribution.
December 17, 2025 at 12:03 PM
byrdnick.com
My mental accounting did not budget for @rthaler.bsky.social endorsing crypto on Twitter.
@R_Thaler I’ve been working in crypto for a while now, and it’s finally paying off. Today I picked up a brand-new Cadillac Escalade. I’m really grateful for the guidance from my coach, @CC_btc_bianca, who helped make it possible. 11:15 AM · Dec 16, 2025
December 16, 2025 at 11:04 PM
byrdnick.com
What predicted inappropriate antibiotic prescribing in Japan's #primaryCare clinics?

- Acute bronchitis, common cold (acute upper respiratory), and rhinitis diagnoses
- Older male clinicians
- Telehealth and regular (vs. after) hours appointments

doi.org/10.1136/bmjp...

#AMR
Figure 1. Association of patient, physician and visit characteristics with prevalence of inappropriate antibiotic prescriptions in solo practice clinics. Table 3 shows the three most frequent diagnoses and antibiotics for each appropriateness category.
December 16, 2025 at 12:09 PM
byrdnick.com
Does #morality of a violation depend on your relationship to the wrongdoer?

People from #China and the #UnitedStates rated transgressions involving parents less immoral than transgressions involving a #sales person or superiors (N > 1200).

doi.org/10.3390/bs15...

#ethics #xPhi
"we investigate whether moral evaluations vary depending on the relationship between the actor and the victim. Unlike previous research that primarily adopts a third-party perspective, this study uses a first-person approach, focusing on judgments made by individuals directly involved in the moral interaction. Three empirical studies were conducted: Study 1 tests the influence of social relationships on moral judgment using Chinese participants; Study 2 explores how moral judgments differ across various moral domains in relational contexts; and Study 3 compares Chinese and American participants to assess cross-cultural differences in the impact of social relationships on moral evaluation. Across all three studies, the results consistently show that social relationships significantly affect moral judgment, supporting the view that moral evaluations are shaped not only by the nature of the act but also by the relational context in which it occurs." Study 1: "Post hoc comparisons using Tukey’s HSD test revealed that moral transgressions involving parents were perceived as the least immoral, whereas those involving a salesperson were judged as the most immoral. The difference between these two groups was statistically significant (p < 0.001; 95% CI = [−0.97, −0.30]). No significant differences were found among the other relational contexts." Study 2: "There was an interaction between social context and moral foundation (F(12, 2424) = 33.00; p < 0.001; partial η 2 = 0.14). The results also showed a main effect for the relational context. Post hoc comparisons using Tukey’s HSD test indicated that moral transgressions involving parents were perceived as the least immoral, whereas those involving a superior were judged as the most immoral. The difference between these two groups was statistically significant (p < 0.001; 95% CI = [−0.34, −0.13]), while no significant differences were found among the other relational contexts." Study 3: "results from the interaction analysis indicated a significant interaction effect between social relationships and culture (F(3, 1716) = 51.88; p < 0.001; partial η 2 = 0.08). Furthermore, both the main effect of social relationships and the main effect of culture reached statistical significance. To further examine the main effect of social relationships, a post hoc analysis using Tukey’s HSD test was conducted. The findings indicated that moral transgressions involving parents were perceived as the least immoral, whereas those involving a salesperson were judged as the most immoral. The difference between these two relational contexts was statistically significant (p < 0.001; 95% CI [−0.34, −0.13]), while no significant differences were observed among the remaining relational categories."
December 15, 2025 at 12:39 PM
byrdnick.com
Only "5% of mass shootings are motivated primarily by #mentalIllness". Of course, "the lifetime prevalence of any psychiatric disorder among adults in [the U.S.] right now is about 40 to 45%, so a lot of people who perpetrate mass murder will have some sort of diagnos[is]."
1/5
December 14, 2025 at 2:46 PM
Reposted by Nick Byrd, Ph.D.
benvancalster.bsky.social
Our guidance regarding performance measures for medical AI models is finally out!

- Stop bashing AUROC, although it does not settle things
- Calibration and clinical utility are key
- Show risk distributions
- Classification statistics (e.g. F1) are improper

www.thelancet.com/journals/lan...
Evaluation of performance measures in predictive artificial intelligence models to support medical decisions: overview and guidance
Numerous measures have been proposed to illustrate the performance of predictive artificial intelligence (AI) models. Selecting appropriate performance measures is essential for predictive AI models i...
www.thelancet.com
December 13, 2025 at 2:04 PM
byrdnick.com
Are more reflective thinkers better at games that require empathy or perspective-taking?

In two experiments, reflective thinking performed better on such games, seemingly because they paid more attention to the other players' incentives.

doi.org/10.1017/S193...

#econ #cogSci
"Nash equilibrium is a game solution in which none of the players has a self-interested incentive to deviate from its own strategy after considering the counterpart’s choice. In Experiment 1, we used a particular class of game called dominance-solvable. These games contain an option which is better than another one for a player, independently of the action the counterpart will take. We refer to this option as a dominant strategy.3 In Experiment 1, we used two classes of dominance-solvable games characterized by different equilibrium structures, creating sixteen 2x2 games for each class (for a full list of game matrices, see Figure A1 in section A.1, Appendices). The two classes of games (Figure 1) were: (1) dominance solvable “self” games (DSS), in which only the participant had a strictly dominant strategy; (2) dominance solvable “other” games (DSO), in which only the opponent had a strictly dominant strategy." "In order to explore the type of visual analysis performed by participants, transitions were divided in five major types (Figure 2), following the classification of Devetag and colleagues (2016): 1) own-payoffs within-action transitions: transitions between player’s own payoffs within a single row (necessary to identify the action with the highest average payoff). 2) own-payoffs between-action transitions: .... 3) other-payoffs within-action transitions: ... 4) other-payoffs between-action transitions: transitions between the counterpart’s payoffs within a single row (necessary to identify the presence of counterpart’s dominant choices). 5) intra-cell transitions: transitions between the payoffs of the two players, within the same cell (necessary to compare the two players’ payoffs given a specific combination of choices)." "the model that best predicts the proportion of equilibrium choices included only the CRT score (R 2 = .11, F (1, 46) = 5.59, B = 0.33, p = .022), while we did not find any effect of fluid intelligence or working memory". "the average causal mediation effect of proportion of other-payoffs within-action transitions on the relation between CRT score and proportion of equilibrium responses is statistically significant (p = .002, based on 10000 bootstrap samples), accounting for an estimated 58 % of the total effect between CRT score and proportion of equilibrium responses (Table 4)."
December 12, 2025 at 12:38 PM
byrdnick.com
In the 60s Wason showed people poorly understand the #logic of testing abstract conditionals (If..., then...).

In the 70s Johnson-Laird et al. found people DO know how others tend to test FAMILIAR conditionals (If someone drinks, then they must be 21).

After many more papers...
"The overall goal of the Wason selection task is to analyze how subjects formulate and test hypotheses from a conditional "If p then q" rule. This problem has been and continues to be one of the best experimental tools for studying the mechanisms underlying human reasoning. The original abstract version of the problem 1 is illustrated in Figure 1." "Thematic" versions of Wason's card selection task which involve familiar (rather than abstract) content.
December 11, 2025 at 12:04 PM
Reposted by Nick Byrd, Ph.D.
journalphp.bsky.social
In #vol38issue2 Miklós Kürthy, Graham Bex-Priestley & Yonatan Shemmer argue that in our common understanding disagreement is not just a clash of beliefs but also a clash of intentions #openaccess doi.org/10.1080/0951... @philsky
4/14
It’s common sense – you don’t need to believe to disagree!
It is often assumed that disagreement only occurs when there is a clash (e.g., inconsistency) between beliefs. In the philosophical literature, this “narrow” view has sometimes been considered the ...
doi.org
February 5, 2025 at 3:11 PM
byrdnick.com
Does open-minded reflection improve #diagnosis?

Reflecting on reasons for, objections to, expected-but-missing symptoms of, and an alternative to medical residents' initial diagnoses improved accuracy, albeit not significantly (p = 0.263, N = 56).

doi.org/10.1007/s104... #medEd
"Participants were randomly assigned to the learning-by-teaching condition (n = 28) or the control condition (n = 28)." "Ninety-nine residents followed our invitation and came to the first study session, and 56 of them (39 female; age M = 29.05, SD = 2.85) completed both sessions." Medical residents randomized to "the experimental condition [studied three] examples of the procedure that they would be asked to teach. ... Each example showed the reflection procedure [which[ requires the physician to first read a case and give an initial diagnosis. Subsequently, the physician goes back to the case and lists (1) findings that support the diagnosis, (2) findings that oppose the diagnosis, (3) findings that would have been expected if the diagnosis were true, but were absent. The physician then gives (4) an alternative diagnosis and follows the same analytical steps (1–4) for this alternative diagnosis, and for a third diagnosis." Table 2 All outcome measures collected during the post-test "One-way ANOVAs showed no main effects of condition on diagnostic accuracy, F (1, 54) = 1.28, p = 0.263". The authors admit that their sample was almost half as large as they had hoped and, thus, perhaps insufficiently powered to detect the effect they expected.
December 10, 2025 at 12:53 PM
byrdnick.com
Can forgetting to take your #medicine be associated with where you store it?

In a survey of over 1500 people, forgetting to take #medication was
- higher if stored in kitchen cabinets
- lower if stored in desks, dining room tables, and nightstand drawers

doi.org/10.3390/bs14...
"The home medication storage locations that were significantly associated with decreased odds of having... forgotten to take a medication included kitchen tables (OR: 0.67, 95% CI: 0.50–0.88)" The "home medication storage locations were significantly associated with having decreased odds of forgetting to take a medication [were] atop bedroom nightstand (OR: 0.73, 95% CI: 0.58–0.91), in the nightstand drawer (OR: 0.60, 95% CI: 0.48–0.75), desk (OR: 0.75, 95% CI: 0.56–0.99), [and] on the dining room table (OR: 0.70, 95% CI: 0.49–0.99)"
December 9, 2025 at 12:22 PM
byrdnick.com
Can it help to reflect on discrepancies between intended and actual #alcohol consumption?

For male undergraduates, this reflection reduced #drinking (compared to a control group).

For females, merely enrolling in the study was enough to reduce drinking.

doi.org/10.3390/bs14...
"Method: Participants (N = 88, females = 61.5%, mean age = 21.05 years) were randomly assigned to a control group or [a] Computerized Brief Intervention (CBI) ...to motivate participants to change their drinking behavior. ... At baseline and a three-month follow-up, the participants completed a battery of measures of alcohol consumption and related problems." "Results: At baseline, the participants were confirmed to be heavy drinkers with many drink-related negative consequences. ... During follow-up, males’ alcohol use was [lower in the treatment groups compared to the control group]. The females in all three groups reduced their alcohol use, but there were no significant group differences." "Accordingly, 90 participants were recruited from students at Bangor University through an offer of either course credit (one hour of research credit for each hour that they participated) or a modest cash payment of GBP 10. The inclusion criteria were that men should drink more than 21 units of alcohol per week or 8 or more units on one occasion at least weekly, and women should drink more than 14 units per week or 6 or more units on one occasion at least weekly. Each UK unit of alcohol represents 10 mL or 8 g of pure alcohol. The participants were randomly assigned to one of the three groups: Computerized Brief Intervention (CBI), Computerized Brief Intervention-Enhanced (CBI-E), or the Control group. The final sample who completed the first session included 88 students, 54 (61.5%) of whom were female, and their mean age was 21.05 years (SD = 4.42). Seventy-five participants completed the follow-up session; 60% of them were female." "The participants [in the treatment groups] were introduced to the Computerized Brief Intervention (CBI) as a computer-assisted interview during which they would receive personalized and objective feedback about their drinking. The CBI aimed to motivate participants to change their current use of alcohol through the use of the following components: First, participants were given objective feedback about their drinking. Second, the discrepancy between participants’ current drinking and their drinking goal was highlighted. Third, participants were asked to consider the implications of not changing their current level of alcohol use. The CBI achieved these objectives in approximately 30 min through the use of 12 computer screens." "For males, after a steady reduction in drinking during the first three weeks after being included in the study, the weekly consumption of each of the groups stabilized. Individuals in the CBI group[s] drank less alcohol than the [control] group." "For females, however, the post-intervention pattern of drinking was quite different. First, the female participants in all three groups considerably reduced their consumption just after they completed the baseline assessment. Second, the reductions were maintained throughout the 12 weeks of follow-up. Third, the differences in consumption among the three groups were less distinct than for the males."
December 8, 2025 at 12:01 PM
Reposted by Nick Byrd, Ph.D.
nadirafaber.bsky.social
In our new paper (open access in PSPB), we asked where children’s greater moral concern for animals stems from. We find that children place less moral weight on species membership alone (i.e. are less speciesist) than adults. #socialpsyc #devpsy #philsky journals.sagepub.com/eprint/ZMRUM...
November 11, 2025 at 1:09 PM
byrdnick.com
Imagine a software error assigned two people to the final #conference speaking slot and you have to decide which person is removed from the program.

Guess which factors predicted whether people #canceled a speaker.

(The survey respondents were 1540 German citizens.)
"In the experiment, participants were confronted with a canceling problem. Specifically, they were asked to imagine that they are responsible for organizing a conference on ophthalmology. Due to a software error, two speakers have been mistakenly assigned to the last available speaking slot, and it is impossible to let both of them present. The task requires the participants to make a difficult decision: one of the two speakers must be removed from the list of presenters. Participants were informed that they had conducted online searches for both professional and personal information about the speakers and had a brief phone conversation with each to confirm their interest in presenting at the conference. They were also told that each speaker had made either a politically charged or a neutral statement, conveyed either during the private phone conversation or through a blog or social media post." "Texts used as framing stimuli in the choice experiment. IMPORTANT: Research has shown that the best decisions are those that are made [intuitively and from the gut] [using logic and reasoned thinking]. In the following, we are interested in your [intuitive] [deliberate] decisions. Therefore, please decide in each case, as [spontaneously] [thoughtfully] as possible, from which of the two sellers you would buy."
December 7, 2025 at 12:42 PM
byrdnick.com
Moderate confidence signaled more expertise than high confidence among more reflective thinkers and "variability in ...confidence serve[d] as an additional cue to expertise".
mauricio19.bsky.social Mauricio Palmeira @mauricio19.bsky.social · Jul 16
Do confident people look more knowledgeable? Not always. We show that high confidence equals more expertise among some, but others have the opposite intuition. We show that more deliberate thinkers (high CRT) generally prefer moderate over high confident judgments.
December 7, 2025 at 4:26 AM
byrdnick.com
Can doctors save time and stress with #AI scribes who listen to interactions with patients and create medical notes?

After many non-randomized studies without control groups, there are now results from two randomized control trials:
"In the more typical case, observed by Lukac et al., AI scribes saved less than a minute per visit, consistent with prior observational studies." Kim, E., Liu, V. X., & Singh, K. (2025). AI Scribes Are Not Productivity Tools (Yet). NEJM AI, 2(12), AIe2501051. https://doi.org/10.1056/AIe2501051
December 4, 2025 at 12:09 PM
byrdnick.com
Just downloaded #Redfin and offloaded #Zillow.
hatethegamebook.com Daryl Fairweather ⛅ @hatethegamebook.com · 26d
You can still find property level climate risk scores on Redfin

Zillow Removes Climate Risk Scores From Home Listings
www.nytimes.com/2025/11/30/c...
Zillow Removes Climate Risk Scores From Home Listings
www.nytimes.com
November 30, 2025 at 7:32 PM
byrdnick.com
Seeking a PhD in Cognitive Science? 👇
davidlevari.bsky.social David Levari @davidlevari.bsky.social · Oct 30
PhD applications for Cognitive and Psychological Sciences at Brown (@browncopsy.bsky.social) are due on December 1st. We have a great community, so please apply, share with your students, etc. Read more on our fancy new website here: copsy.brown.edu/graduate-study
Graduate Study
Our graduate programs are highly selective, successfully recruiting and retaining excellent students, competing with the top programs in the world. Our doctoral students go on to be scientists and sch...
copsy.brown.edu
November 24, 2025 at 12:54 AM