Sign in

Matthew D Johnston

@mattdjohnston.bsky.social
104 followers 205 following 23 posts

Lecturer at York St John | Research interests include the development of mental time travel skills, counterfactual emotions, and statistical learning. scholar.google.co.uk/citations?user… I also play bass and drink too much tea ☕

PostsRepliesMedia
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
8) Thanks to all the children who took part in these (quite repetitive) games and to the mammoth amount of work Jessica Dolan put into organising the sessions. Also, thanks again to the RAs who helped with collecting the data.
001
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
7)This suggests young children may be especially good at APPLYING previously learned statistical info to new contexts. Regarding SL-EF links, we found little evidence SL drives EF development, though more work is needed with younger kids and auditory tasks.
101
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
6) UNEXPECTED: In one task, despite introducing novel alien-triplet stimuli each time point, the 5-6-year-olds showed the greatest longitudinal gains in sensitivity to the statistical regularities.
101
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
5) RESULTS: Children showed statistical learning in all three SL tasks. In two of three SL tasks they became increasingly sensitive to statistical regularities across sessions (i.e., with 8 month gaps).
101
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
4) METHODS: Using an accelerated longitudinal design, we tested the same children (5-6-, 8-9-, and 11-12-year-olds) from our previous article testing their SL and EF abilities at three time points over 16 months.
101
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
3) Higher-order EFs develop via brain maturation, but children also accumulate goal-relevant cues with age. We theorised early SL abilities help build these cues, driving EF gains.
101
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
2) Here we had 2 aims: 1) explore how statistical learning (SL) abilities change over time, and how previously learned patterns affect later learning; 2) test the relation between SL and higher-order cognitive abilities (working memory, inhibition, set shifting) that support goal-directed behaviour.
101
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
1)BACKGROUND: This is the sister paper to our previous article on children's ability to learn different types of statistical patterns in their environment (e.g., infants learning when certain syllables predict each other in speech, and how often these sounds occur).
101
Matthew D Johnston @mattdjohnston.bsky.social · 17/08/2026
My 2nd postdoc paper has been published @srcdorg.bsky.social 🎉 This was a lot of work for us and the kids taking part (6 sessions, 16 months)! Huge thanks to co-authors, Nic Chevalier, Jessica Dolan, and Bence Farkas and to all the RAs that helped collect the data! doi.org/10.1093/chid... See🧵:
110
Matthew D Johnston @mattdjohnston.bsky.social · 17/04/2026
Thanks to all volunteer RAs helping to collect the data! Also, a massive thank you to the families bringing thier children to the labd and to the children for taking part in these (highly repetitive) games!!
000
Matthew D Johnston @mattdjohnston.bsky.social · 17/04/2026
6/6 Performance across the 3 tasks was unrelate supporting the idea that SL isn't one unified ability but has multiple distinct components. Rsearchers should be clear about the type of patterns participants learn in their tasks. We also need tasks with higher reliability with child populations.
100
Matthew D Johnston @mattdjohnston.bsky.social · 17/04/2026
5/6 We also looked at executive function (EF). Despite some theories suggesting immature top-down control helps SL, we found only weak evidence of a negative association between EF and SL, and only on the novel alien task (and not the tasks assessing cue-based or probabilistic learning).
100
Matthew D Johnston @mattdjohnston.bsky.social · 17/04/2026
4/6 No age differences emerged for cue-based or probabilistic tasks. For adjacent sequences, younger children showed greater overall sensitivity (i.e., fast for predictable trials). Note though the *rate* at which children picked up these patterns did not change with age.
100
Matthew D Johnston @mattdjohnston.bsky.social · 17/04/2026
3/6 Take our alien task as an example. Children "helped aliens find their spaceships" by pressing a key based on how many eyes each alien had. Hidden in the task was a repeating order in which the aliens appeared (triplets). Learning was measured from RTs to predictable vs unpredictable aliens.
100
Matthew D Johnston @mattdjohnston.bsky.social · 17/04/2026
2/6 187 children completed 3 tasks, each targeting a different type of pattern: (1) cue-based patterns (a cue signals something important); (2) probabilistic sequences (some events occur more than others); and (3) deterministic adjacent sequences (events that co-occur in a fixed order).
100
Matthew D Johnston @mattdjohnston.bsky.social · 17/04/2026
1/6 Young children are impressive learners. They pick up repeated patterns in the environment without being taught (e.g., infants learn which sounds tend to go together in speech). This is called statistical learning (SL), and we wanted to know how it develops across 5-12-year-olds.
100
Matthew D Johnston @mattdjohnston.bsky.social · 17/04/2026
New @srcdorg.bsky.social paper exploring links between statistical learning and executive functions in middle childhood! Link here: doi.org/10.1093/chid... w Nic Chevalier, Jess Dolan, and @nemeth.bsky.social See thread below...
doi.org
Statistical learning in childhood: Dimensions, developmental trajectory, and relation with cognitive control
Abstract. Compared to other developmental processes, such as cognitive control, there is relatively little consensus on the developmental trajectory of sta
141
Matthew D Johnston @mattdjohnston.bsky.social · 31/07/2025
Looking forward to presenting our research tomorrow at this family-friendly event - a great opportunity to learn about cognitive research at the #WeeScience lab. 📅 Tomorrow 10AM - 3PM (drop in) 📍 Bayes Centre ✅ Free & open to all Don't miss this day of discovery! Register: tinyurl.com/CogSciDay
010
Matthew D Johnston @mattdjohnston.bsky.social · 28/05/2025
Very belated post...Had an amazing time presenting Carnegie Trust work w @kwolfe.bsky.social at #EUCAS2025 conference at Università di Pavia. Beautiful venue plus we were greeted with amazing risotto 🍽️ Learned alot about ageing research and made great connections, including folks from #SCAN!
010
Matthew D Johnston @mattdjohnston.bsky.social · 06/05/2025
✈️ I'm still recovering from jetlag but had a great time at #SRCD2025! Thrilled about our symposium on cognitive control and statistical learning. @nemeth.bsky.social and Rachel Foster gave fantastic talks—Rachel did amazingly well for her 1st conference and will soon be on the job market!
031
Matthew D Johnston @mattdjohnston.bsky.social · 02/05/2025
Poster at 1640 today showcasing a study where we looked at whether increasing cognitive load in a sorting task impacted children and adults' ability to learn statistical regularities. If for nothing else come see my improving photoshopping skills on display... Exhibit hall B, Poster #5 #SRCD2025
020
Matthew D Johnston @mattdjohnston.bsky.social · 01/05/2025
Excited to be opening for this symposium on developmental dynamics between cognitive control and statistical learning today! @nemeth.bsky.social, Rachel Foster, and Qianqian Wang. Time: 10:00 AM - 11:30 AM 203AB | Minneapolis Convention Center, Level Two #SRCD2025 #statisticallearning #ef
170
Matthew D Johnston @mattdjohnston.bsky.social · 01/05/2025
Excited to be opening for this symposium today with @nemeth.bsky.social, Rachel Foster, and Qianqian Wang. Join us today at 10:00AM in Room 203AB #SRCD #statisticallearning #cognitivecontrol
021