# Research: sources behind the testimony and the post Back to [[Reverse Flynn Effect and EdTech]] · [[Senate Testimony - Horvath (2026-01-15)]] · [[Vocabulary]] Key: ✔ I opened or confirmed the source (DOI, publisher or official page) this session. ◻ The citation is listed as given in the testimony and I did not open it. ⚠️ The detail is unconfirmed. ## A. The 18 references in Horvath's written testimony | # | Reference (as cited) | What it is used for | Check | Notes | |---|---|---|---|---| | 1 | Trahan, Stuebing, Fletcher & Hiscock (2014). "The Flynn effect: a meta-analysis." *Psychological Bulletin* 140(5), 1332. | Rising cognition in the 20th century | ◻ | The standard meta-analysis of Flynn gains (roughly 2.3–3 points per decade). It documents the **rise**, not a reversal. | | 2 | Mullis et al. (2023). *PIRLS 2021 International Results in Reading.* [doi:10.6017/lse.tpisc.tr2103.kb5342](https://doi.org/10.6017/lse.tpisc.tr2103.kb5342) | Decline since the mid-2000s | ◻ | 4th-grade reading. Results were affected by the pandemic. | | 3 | OECD (2023). *PISA 2022 Results (Vol. I).* [doi:10.1787/53f23881-en](https://doi.org/10.1787/53f23881-en) | Decline | ◻ | OECD averages of 472 in math and 476 in reading were then record lows. PISA 2025 has since gone lower (see C). | | 4 | Andrzejewski, Zeilinger & Pietschnig (2024). "Is there a Flynn effect for attention? Cross-temporal meta-analytical evidence for better test performance (1990–2021)." *Personality and Individual Differences* 216, 112417. | Cited for decline in "attention" | ◻ ⚠️ | **The title reports *better* attention test performance from 1990 to 2021.** According to IQ Metrics' reading of the abstract (287 samples, 21,291 people): adults gained moderately, while children did not gain and made more errors alongside faster processing. Overall, the paper only weakly supports "attention declined." | | 5 | Kim, K. H. (2021). "Creativity crisis update." *Roeper Review* 43(1), 21–41. | Decline in creativity | ◻ | Torrance creativity scores. This is not IQ. | | 6 | Dutton, van der Linden & Lynn (2016). "The negative Flynn Effect: A systematic literature review." *Intelligence* 59, 163–169. | Reversal | ◻ | Reviews reversal reports, mainly from Nordic and Western European countries, the UK and Australia ⚠️ (country list from memory, not re-read). | | 7 | Fittes (2022). "How much time are students spending using EdTech?" *EdWeek Market Brief*, Mar 1, 2022. | "Over half… one to four hours" of school computer use | ◻ | A trade-press survey, not peer-reviewed research. | | 8 | Ragan, Jennings, Massey & Doolittle (2014). "Unregulated use of laptops over time in large lecture classes." *Computers & Education* 78, 78–86. | "Off-task for up to 38 minutes of every hour" | ◻ | ⚠️ This is a study of **university lecture** classes, applied here to K–12. | | 9 | OECD (2016). *PISA 2015 Results (Vol. I).* [doi:10.1787/9789264266490-en](https://doi.org/10.1787/9789264266490-en) | Test-mode effects | ◻ | | | 10 | Salmerón, Vargas, Delgado & Baron (2023). *Reading and Writing* 36(1), 175–194. | US PIRLS: digital use vs reading | ◻ | | | 11 | Hattie (2023). *Visible Learning: The Sequel.* | The +0.42 benchmark | ◻ | See [[People/John Hattie]] and the critique in [[Related Research and Trends#Criticism of Horvath's case]]. | | 12 | Horvath (2026). *The Digital Delusion.* LME Global Press. | Meta-analysis tables and mode effects | ◻ | His own book, self-published. Chalkbeat says a major publisher was set to re-release it in summer 2026. Penguin Random House lists ISBN 9798217423965. | | 13 | Kirschner & De Bruyckere (2017). "The myths of the digital native and the multitasker." *Teaching and Teacher Education* 67, 135–142. | Mechanism | ◻ | | | 14 | Jolicoeur, Dell'Acqua & Crebolder (2000). Attentional blink and PRP chapter, MIT Press. | Task-switching cost | ◻ | | | 15 | Wu & Liu (2008). "Queuing network modeling of the psychological refractory period." *Psychological Review* 115(4), 913. | Interference | ◻ | | | 16 | Foerde, Knowlton & Poldrack (2006). "Modulation of competing memory systems by distraction." *PNAS* 103(31), 11778–11783. | Distraction shifts learning to habit memory | ◻ | A lab study in adults. | | 17 | Bergman et al. (2014). "High IQ in early adolescence and career success in adulthood." *Research in Human Development* 11(3), 165–185. | National implications | ◻ | | | 18 | Wraw, Deary, Gale & Der (2015). "Intelligence in youth and health at age 50." *Intelligence* 53, 23–32. | National implications | ◻ | | **What the reference list shows:** - None of the 18 references measures **Gen Z's IQ against their parents' generation.** - Refs 1 and 6 are about the Flynn effect in general. - Refs 2, 3 and 5 are about achievement and creativity. - Ref 4 appears to point the other way on attention. - The "first generation… general IQ" claim rests on the oral statement alone. ## B. Primary research on the Flynn effect and its reversal | Study | Data | Finding | Link | Check | |---|---|---|---|---| | Flynn (1984). "The mean IQ of Americans: Massive gains 1932 to 1978." *Psychological Bulletin* 95(1), 29–51 | US norm comparisons | About 3 points per decade | — | ◻ (citation from memory ⚠️) | | Flynn (1987). "Massive IQ gains in 14 nations." *Psychological Bulletin* 101(2), 171–191 | 14 countries | Large gains, especially on Raven-type tests | — | ◻ (citation from memory ⚠️) | | Pietschnig & Voracek (2015). "One century of global IQ gains: A formal meta-analysis of the Flynn effect (1909–2013)." *Perspectives on Psychological Science* 10(3), 282–306 | 271 samples, about 4 million people | Gains of about 2.8 points per decade, slowing in recent decades | [doi:10.1177/1745691615577701](https://doi.org/10.1177/1745691615577701) | ◻ (DOI from memory ⚠️; numbers as reported by iqmetrics.org) | | Teasdale & Owen (2005). *Personality and Individual Differences* 39, 837–843; (2008) *Intelligence* 36, 121–126 | Danish conscripts | Peak in the late 1990s, then a decline of about 1.5 points by 2003–04 | — | ◻ ⚠️ | | **Bratsberg & Rogeberg (2018).** "Flynn effect and its reversal are both environmentally caused." *PNAS* 115(26), 6674–6678 | **736,808** Norwegian conscript scores, men born 1962–1991, **including brothers** | IQ rose until the cohorts born about 1975, then fell. The decline appears **within families**, so it is environmental, not demographic or dysgenic. The study does not identify the cause. | [doi:10.1073/pnas.1718793115](https://doi.org/10.1073/pnas.1718793115) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC6042097/) (PMC link not opened) | ✔ | | Dutton, van der Linden & Lynn (2016). *Intelligence* 59, 163–169 | Review | The negative Flynn effect in several developed countries | [doi:10.1016/j.intell.2016.10.002](https://doi.org/10.1016/j.intell.2016.10.002) | ◻ (DOI from memory ⚠️) | | **Dworak, Revelle & Condon (2023).** "Looking for Flynn effects in a recent online U.S. adult sample: Examining shifts within the SAPA Project." *Intelligence* 98, 101734 | 394,378 US adults online, 2006–2018 | Declines in overall ability and matrix/series reasoning. **No meaningful change in verbal reasoning.** 3-D rotation rose. The authors warn the sample is **not representative**. | [doi:10.1016/j.intell.2023.101734](https://doi.org/10.1016/j.intell.2023.101734) | ✔ | | **Trudel, Winter & Kaufman (2025).** "Digging In: Uncovering the Subtest and Composite Level Flynn Effect on the WAIS-5." *Journal of Pediatric Neuropsychology* 11(2), 39–48 | WAIS-5 norm comparisons | For 16-year-olds: fluid reasoning +3.0 and processing speed +2.1 points per decade; verbal comprehension −1.2 and working memory −1.8 per decade | [doi:10.1037/jpn0000009](https://doi.org/10.1037/jpn0000009) | ✔ (title, journal and DOI confirmed; per-decade numbers from a search summary of the paper ⚠️) | | **Lesigang, Vetter, Voracek & Pietschnig (2026).** "No Age of Inattention…" *Intelligence & Cognitive Abilities* (Oct 2026) | Austrian adults, 2000–04 vs 2025 (N = 410) | Selective attention slightly better (d = 0.12). Interference control stable. Sustained attention slightly lower (d = −0.27, possibly a hardware artifact). **No sweeping decline in attention.** | [ICA journal](https://icajournal.scholasticahq.com/article/171910-no-age-of-inattention-small-and-domain-differentiated-cross-temporal-changes-in-tests-of-selective-attention-and-attentional-control-in-austrian-adul) | ✔ (abstract via search) | **Horvath's own account of the IQ data** ([Substack, Feb 6, 2026](https://thedigitaldelusion.substack.com/p/i-told-the-senate-gen-z-is-less-cognately)) ✔: - Gains of "approximately three points per decade" in the 20th century. - A negative Flynn effect "as much as six to eight points per decade" across "a handful of well-designed studies." ⚠️ He does not name these studies, and this rate is far larger than the Norwegian estimate. - He leans on "Digging In": performance on internally scaffolded (verbal and working-memory) tasks fell while visual tasks rose. - He concludes: **"For that reason, we cannot concretely say that Gen Z is less intelligent than Millennials… Gen-Z is losing school-ability."** ## C. Achievement data (the strongest part of the "decline" story) | Source | Finding | Link | Check | |---|---|---|---| | **PISA 2025** (OECD, released Sep 8, 2026) | Lowest OECD averages ever recorded in math and reading. Reading fell 28 points from 2015 to 2025 ("about one and a half years of learning") and math fell 22 points. Science was stable from 2022 to 2025. One in five students is a low performer in all three subjects. "Students tended to perform less well when [device] use became excessive or when devices were used for leisure at school." | [OECD press release](https://www.oecd.org/en/about/news/press-releases/2026/09/pisa-2025-students-reading-and-mathematics-performance-declined-sharply-across-the-oecd.html) | ✔ | | **NAEP 2024, 12th grade** (released Sep 9, 2025) | Math and reading each fell 3 points from 2019, both the lowest ever reported. Lower percentiles fell the most. | [NAGB](https://www.nagb.gov/news-and-events/news-releases/2025/declines-in-8th-grade-science-and-12th-grade-math-and-reading.html) | ✔ (via search summary) | | Twenge (2026). "International declines in academic performance and increases in loneliness are linked to electronic devices." *Journal of Adolescence* 98, 250–261 | PISA: science declines were steeper where students used devices more for leisure at school | cited in her testimony | ◻ | | OECD (2015). *Students, Computers and Learning* | Heavy school computer use was linked to lower PISA scores | ⚠️ (I did not open it; it is likely the basis for "80 countries… two-thirds SD," but this is unconfirmed) | ◻ | ## D. Mechanism claims | Claim | Source | Check | |---|---|---| | Speaking (in person or by phone) releases oxytocin, while instant messaging does not | Seltzer, Prososki, Ziegler & Pollak (2012). "Instant messages vs. speech: hormones and why we still need to hear each other." *Evolution and Human Behavior* 33(1), 42–45. [doi:10.1016/j.evolhumbehav.2011.05.004](https://doi.org/10.1016/j.evolhumbehav.2011.05.004) | ✔ (a small study of girls aged 7–12 after a stressor; authors from memory ⚠️) | | Text-based communication releases "tachykinins… precursor to chemical depression" | No source found | ❓ | | Paper beats screens for reading comprehension | Testimony ref 12. Also Delgado et al. (2018) meta-analysis, *Educational Research Review* 25, 23–38 ⚠️ (from memory) | ◻ | | Handwritten notes beat laptop notes | Mueller & Oppenheimer (2014), *Psychological Science* 25(6) ⚠️. A replication by Morehead, Dunlosky & Rawson (2019), *Educational Psychology Review*, found the advantage was **not robust** ⚠️ (from memory) | ◻ | ## E. Secondary coverage used - **Chalkbeat**, Matt Barnum, Mar 17, 2026: ["A viral case against screens in schools is winning converts. Does the evidence hold up?"](https://www.chalkbeat.org/2026/03/17/jared-cooney-horvath-says-ed-tech-hurts-learning-a-look-at-the-evidence/) ✔ - **Fortune** (both URLs checked by page title ✔): - ["Neuroscientist warns Gen Z first generation less cognitively capable than their parents"](https://fortune.com/article/how-did-us-spending-30-billion-dollars-on-laptops-result-in-first-generation-less-cognitively-capable-than-parents/). According to IQ Metrics, it attributes the "first generation" line to the *written* testimony, the same error as the post. - ["Neuroscientist Jared Horvath: Silicon Valley pushed computers in classrooms, plummeting test scores"](https://fortune.com/article/american-schools-silicon-valley-lie-edtech-chromebook-ai-back-to-school/). - **Education Next**, Carl Hendrick, Feb 10, 2026 (26(1)): ["A Latter-Day Luddite Pulls the Plug on EdTech"](https://www.educationnext.org/latter-day-luddite-pulls-the-plug-on-edtech-book-review-digital-delusion-jared-cooney-horvath/) ✔. - A largely favorable review. It gives the book's **raw** effect sizes: overall EdTech +0.29 (398 meta-analyses, more than 21,000 studies), intelligent tutoring systems +0.52, learning disorders +0.61, 1:1 laptops +0.16, disadvantaged students +0.18. - These confirm that the testimony's negative numbers are re-centered against the +0.42 benchmark. The 1:1 figure differs slightly: the testimony gives −0.30, which would imply a raw effect of +0.12, while the review gives +0.16 ⚠️. - The review also cautions: "attributing it primarily to education technology… requires more careful causal analysis than meta-analytic correlations can provide." - **iqmetrics.org** (Oct 10, 2026) [Gen Z IQ explainer](https://iqmetrics.org/news/gen-z-iq-is-gen-z-less-intelligent-than-its-parents-what-the-data-show/). A secondary aggregator, used for leads; I checked its key claims against the primary sources above. - It found that Norwegian IQ peaked for men born in 1975 at 102.3, fell to 99.4 for the 1989 cohort, and declined about 0.33 points per year within families (roughly 3.3 per decade). **Those data stop with the 1991 birth cohort, six years before the first Gen Z birth year** (Pew: 1997). - It also cites Winter, Trudel & Kaufman (2024), "Wait, where's the Flynn effect on the WAIS-5?" (*Journal of Intelligence*), which found a smaller gain of 1.2 points where about 3 were expected, not a loss ◻. - It says it "could not identify" the "Digging In" paper. I did find it; see B above.