David Melnikoff

@DEMelnikoff

Social Psychology Postdoc @ Northeastern University

Vrijeme pridruživanja: siječanj 2018.

Tweetovi

Blokirali ste korisnika/cu @DEMelnikoff

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @DEMelnikoff

  1. 29. sij

    So, if you want to know more about the theoretical and empirical case for furnishing models of action control with what we know about the format of action representations, this is a great place to start.

    Prikaži ovu nit
    Poništi
  2. 29. sij

    This basic idea parallels recent work in RL by folks like , () & Botvinick (), who, under the heading of Episodic RL, have started appealing to representation to address various computational challenges.

    Prikaži ovu nit
    Poništi
  3. 29. sij

    In this paper, Hommel highlights what he sees as major shortcomings in dominant (process-focused) theories of habit, and suggests that the Theory of Event Coding offers some promising solutions....

    Prikaži ovu nit
    Poništi
  4. 29. sij

    Specifically, it explains action control in terms of the modal, distributed nature of action representations, rather than specific learning or decision-making algorithms. It happens to be uniquely capable of explaining many classic phenomena, including Stroop and Simon effects.

    Prikaži ovu nit
    Poništi
  5. 29. sij

    Hommel & colleagues developed a highly influential and empirically successful theory of action control called the Theory of Event Coding (). It's a unique theory in that it emphasizes representational format over process...

    Prikaži ovu nit
    Poništi
  6. 29. sij

    I highly recommend that anyone interested in action control read this paper by Bernhard Hommel (). Anyone interested in Episodic RL who hasn't read Hommel's work will find it especially interesting 👇

    Prikaži ovu nit
    Poništi
  7. 25. sij

    Now, about those RL papers. For gentle introductions I'd recommend, in addition to the M&C paper here, the following: , , & , & if you want to go whole hog, this is the canonical text:

    Prikaži ovu nit
    Poništi
  8. 25. sij

    Morris & Cushman present the steel man versions of Dezfouli & Balleine's arguments and data, and respond with their own studies, which they contend support aspects of the D&B model, while at the same time preserving an important role of MF RL in habit formation. A must read!

    Prikaži ovu nit
    Poništi
  9. 25. sij

    The paper presents *very* clever experiments that address an important issue: the putative alignment between habits and model-free RL. Though habits are widely thought to emerge from MF algorithms, Dezfouli & Balleine (e.g., ) have challenged this idea...

    Prikaži ovu nit
    Poništi
  10. 25. sij

    If you've heard about computational RL and wanna know what all the fuss is about, but find the math scary, this paper (plus w/ some others I'll share at the end of this thread) is a great place to start. And that's just a bonus...

    Prikaži ovu nit
    Poništi
  11. 25. sij

    ...starting with this one . Led by Adam Morris & of the , it offers, among other things, a beautifully written, crystal clear articulation of some core concepts, methods & current debates in the computational RL literature.

    Prikaži ovu nit
    Poništi
  12. 25. sij

    Let me first say that the authors wrote some truly excellent papers that triangulate on some exciting themes, such as exploring more collaborative models of habitual & goal-directed control, & re-centering the “Law of Practice” in theories of habit formation. I'd read them all...

    Prikaži ovu nit
    Poništi
  13. 25. sij

    I recently had the pleasure of helping Wendy Wood and John Bargh put together a collection of papers all about habit and RL (). Over the next week or so, I’ll be discussing the virtues of each paper in a series of explainer threads, starting w/ this one 👇

    Prikaži ovu nit
    Poništi
  14. proslijedio/la je Tweet
    Poništi
  15. proslijedio/la je Tweet

    We’re hiring! The IASLab is seeking applications for a Research Technician position to work on an NIH-funded project examining how motivational and affective factors influence cognitive aging. Interested applicants should email c.shaffer@northeastern.edu. Please spread the word!!

    Poništi
  16. proslijedio/la je Tweet
    23. sij

    What an opportunity to work with one of the best scientists/mentors in the biz...

    Poništi
  17. 22. sij

    11/n Hopefully I’ve enticed you enough to read our paper! If so, let us know what you think 😊

    Prikaži ovu nit
    Poništi
  18. 22. sij

    10/n including the pressing question of why the relationship between implicit attitudes & behavior is so unpredictable (see et al.: ). We also explain how prepared reflexes could quickly, easily, & substantially reduce implicit bias.

    Prikaži ovu nit
    Poništi
  19. 22. sij

    9/n As we explain in our paper, prepared reflexes change attitudes in ways not predicted by any existing theory of attitude change, making them uniquely capable of resolving some longstanding puzzles…

    Prikaži ovu nit
    Poništi
  20. 22. sij

    8/n Across 6 studies we found that prepared reflexes indeed change attitudes, and powerfully so — as I eluded to at the start, prepared reflexes were even capable of completely eliminating implicit negativity towards Hitler. Why does this matter?

    Prikaži ovu nit
    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·