The Maddox article was a different sort of view on replication from what we have seen thus far, in my opinion. While many of the articles we have gone over and presentations we have seen seem to have primarily focused on the issues and controversies when findings are not replicated, this article beautifully seams together both studies that have and have not replicated and how this supports or does not support theories. In part, this is because this article is a review and not a study, but I still thoroughly enjoyed the set-up and transitioning throughout the paper.
I am curious how studies/articles were chosen to be included in this review, though there is no apparent reason presented in the paper to believe selection was heavily biased. To me, the reasoning behind how the theories presented were and were not supported was soundly defended.
The implications for the “real world” seem to be particularly relevant for this review, as this is a phenomenon that is seen in classrooms and outside of the laboratory as explained in the article. With a better understanding of the mechanisms behind spacing effects, things like workshops, training sessions, classroom materials, etc. can be better designed to take advantage of these mechanisms. As well, with a better understanding of theory can come better research design to examine this effect further.
This area this week’s article focuses on is very new to my knowledge. It contains a lot of information about the research that has been done on spacing effect. The author first pointed out the limitation of meta-analytic approach in the studies of spacing effect, which is “the selected search and inclusion criteria can limit the scope of the review”. Then, the author posited his solution to address this limitation by examining critical findings in spacing effect literature that have been absent in prior review and assessing past and present theoretical accounts based on the consistent critical findings. This approach of reviewing findings first, then evaluating the extent to which prior theories can accommodate the findings is very different than meta-analysis studies I have read so far. I am wondering whether this way of review can be replicated to other field of knowledge to address the limitations of a meta-analytic approach.
Generally speaking, since long-term benefit is the goal, so a lot of studies reviewed in this article focus on how spacing effect influence final test performance. I am thinking whether the critical findings or different account can be consistent in different domain of learning in terms of conceptual replication. In addition, it seems that the combined encoding variability and study-phase retrieval mechanism can accommodate all consistent findings in the literature. The author posited that when encoding variability was too much, it may lead to a failure of study-phase retrieval, the question I have here is that whether the variability is too much or not is domain specific. In conclusion, for future replication, it will be interesting to examine spacing effect on different domains, different forms of learning tasks, and different level of learning tasks.
Coming into reading this article, I thought that I was fairly informed when it came to the spacing effect. As I read this article I learned that I was very wrong ;) As a review article, I felt that this paper did a fantastic job of covering all of the nuisances of the spacing effect and I particularly enjoyed how each section devoted time to tying back to previous sections to allow the reader to get a more clear picture as to how all these puzzle pieces may fit together. Each section built very nicely on each other—though I felt that some of the sections got a bit bogged down in jargon and made it a bit more difficult to understand how some of the different pieces fit together.
I don’t want to spoil to much before Alex and I lead discussion but I was particularly intrigued by how the breakdown of the specific mechanisms helped to better explain the spacing effect as a whole. I had not even considered incidental encoding in regard to the spacing effect before—though the phenomena itself is an interesting one. One other facet that I was also most intrigued by was the discussion of lag and how that effects encoding and retrieval. I mean, after I read the breakdown it made so much more sense that lag interval would have a significant effect on spacing and retrieval success % but I never even considered that as a factor before (I’ll save the rest of what I have to say for discussion).
As for replication, the spacing effect is interesting because it has already been replicated countless times for many, many years. I know the spacing effect is often seen in memory research but I wonder if the we may see the benefits of the spacing effect transfer to other domains—perhaps domains that require procedural memory and countless repetitions such as shooting a basketball. Would these types of memories qualify?
This week reading is interesting. My first impression about this article is the article is not a good example of replication. The author did nice job putting all the information and theories to support their argument but did not mention how replicated with the current research questions that the author looking for. Spacing effect refers to the mnemonic benefit of spacing repeated study events across time compared to massing. The spacing effect was first reported over 100 years ago and refers to the long-term memory benefit produced by spaced study events compared to consecutive study events. The benefit of spaced study is further modulated by the specific interval (i.e. lag) that separates presentations of an item (i.e. the lag effect). Spacing effect research has been subject to numerous reviews which have been useful in developing and evaluating theoretical mechanisms proposed to account for the phenomenon. But at the end of the day I feel like how much people care about spacing effect right now in this technological advance world The article also states two objectives to better evaluate the mechanisms proposed to underlie the spacing effect, 1) critical findings in the spacing effect literature that have been absent in prior reviews, 2) consideration of various instantiations of the encoding variability account which is beyond the basic assumptions. That provides as a means of assessing the viability of an encoding variability mechanism to explain the long-term memory (LTM) benefit of spaced stud. Methodology wise author did nice job fulfill this two objective very thoroughly. The author did also mentioned that consistent findings from the extant spacing effect literature that had not previously been considered in concurrent fashion, and in turn this allowed for further specification of the theoretical mechanism underlying the spacing effect. The study-phase retrieval component of the underlying mechanism showed the individual and group differences will be necessary considerations when designing spaced study schedules. One of the limitations of the literature noted above suggest that there is still substantial opportunity to parameterise and examine the spacing effect. Overall, I enjoyed this week reading !
I really enjoyed this week’s article. It was easy to read, and it was interesting to learn about all of the different aspects of the spacing effect that have been investigated. There were examples of both conceptual and exact replications, which I appreciated, considering the theme of the class. One thing that I thought was interesting was that the authors kind of talked down on meta-analyses, and how certain articles can be left out if they don’t support theories, but then I’m sure they had to pick and choose articles to include or exclude in their own review.
My overall takeaway from this reading is that replication can really help theories develop. An experiment might have extremely interesting results that challenge previous research, but direct replications are needed to figure out whether those results are valid. It might be that researchers have uncovered the exception, not the rule. We know so much about the spacing effect because it has been known about and studied for over 100 years, but there are still unanswered questions, which this article really made me think about. As we learn more, research can be better translated to classroom settings, and learning can be optimized.
This article was a bit technical for me, so most of my commentary consists of questions and confusions.
In the section on experimenter induced variability across sections, one condition described is a cue word change that biases the same meaning, e.g., flower-bulb and tulip-bulb. But these do not have the same meaning and, in general, it can be difficult to determine a word's semantic meaning. There are a few ways to describe or determine meaning, including extension, intension, use. On none of these are these two words the same (though in all cases they partially overlap). My concern is for other word pairs that may have been less similar but deemed the same.
I don't understand the hypothesis formulated by Vorkoeijen and Delaney, that "fast presentation rate would limit spacing between repetitions of massed items in covert rehearsal, whereas a slower presentation rate would allow for longer spacing intervals between covert repetitions of massed items." I suspect my issue is one of vernacular, though it might be my experimental ignorance.
Another point of vernacular: What is a rehearsal? Just the speaking aloud of the words in order, of one word? During presentation? After?
As someone who has used flashcards for language learning, the possibility that massed items produce a sense of knowing (but not necessarily actual knowing!) and thus leading to a decision not to rehearse massed items (at least as much as spaced ones) rings true. Of course, using this as an explanation assumes "control," and the appearance or sense of control is not necessarily control.
What is "qualitative" about the qualitative automatic deficient processing account?
Again, I enjoyed this week’s article. I personally have not read a lot of review articles and this one was great. I think it was very detailed and comprehensive, which is important of course. I am familiar with spacing effects but I would not consider myself to be extremely knowledgeable in this area, so I really appreciated the way the background information was presented. I do have to say, some parts were more difficult to “get through” than others, perhaps that’s just me.
I did get the sense that they were not exactly “for” doing a meta-analysis. After the recent meta-analysis, we discussed in class, I find it almost essential for researchers to explain what criteria was used, despite it being an actual meta-analysis or a review. This criterion allows us to have a better understanding of where the researcher is coming from, and I felt these details could have been included.
I’ve only ever seen spacing effects in research that involves memory, more specifically learning and education. I’m not sure where or why the thought of “would motivation play a role here” came into my mind, but I wonder how motivation could fit into this. Naturally, we think that yes, we reward someone after every time they study, they will continue to study and do better on the test. But do we know this for sure? Are there certain patterns in rewards or motivation that are more effective than others in terms of spacing effects? For example, if we were to have two groups of students with an upcoming exam, one who received extra credit in a course where they completed spaced online study actives and the other who did not, would we see a difference in motivation between the two groups? Or a difference in their performance overall?
This article was primarily interesting because of the research I do with Phil Pavlik and his ITS MoFaCTS. In short, MoFaCTS (Mobile Fact and Concept Training System) is an ITS that spaces concept/fact repetitions to optimize retention and remembering. I also thought I had a relatively solid understanding of the spacing effect, but now I am not so sure.
I wonder if this article delved too deeply? If we say that the spacing effect only works in very (VERY) specific circumstances, is there really an effect? I, for one, think there is a solid effect when items are spaced optimally....optimally being the key word there. It seems pretty straightforward to say that spaced practice inherently requires some sort of deeper remembering process than massed practice. That, in and of itself, seems beneficial to me. Although I am slightly biased, I am a big proponent of the spacing effect and its potential use in classroom/educational settings. Even if we are looking at spacing in a short-term memory way, spacing items for practice will inevitably keep that item in your short term memory "register" in a way that massed practice will not.
This week's article seems to be less directly related to replication than some of our previous readings in some senses. While it largely a review of the literature (certainly an enlightening one), it doesn't really fit into our ideas of directly or even conceptually replicating. What is does do is play an important role in leading the conversation.
The important place for reviews such as these is that distill the conversation. This review was able to concentrate on consistent findings that already exist in the literature on spacing effects. Having a singular, well-written piece that relays the state of the field can be invaluable to researchers looking to replicate findings and advance the field. This particular piece helps to direct scarce resources into possibly enlightening investigations.
This was a great spacing effect paper. However, this review also did a fantastic job of demonstrating the Dunning-Kruger effect. That is, I began this paper thinking my peripheral exposure to the spacing effect literature through different classes and articles on memory was sufficient to help me understanding what it was. Now I realize I only know enough to fool myself into thinking I understand it at all. But, I digress. It was very interesting to see how the author used such fine-grained and nuanced findings to argue against a recent theory. I am not well-versed in the literature to know which explanation is preferred (or even if much thought is regularly given to the mechanisms underlying this effect), but I thought this paper did a good job illustrating the weaknesses of this “reminding” theory. However, I am curious about one of the findings raised as a limitation. The author discusses experimenter-introduced variability across repetitions in one section. Eventually, this leads into a discussion of several studies showing a reduced spacing effect for spaced items paired with different cues. Specifically, this is about the study where they showed a decreased effect in cases where homographs were paired with the same or different cues across conditions (e.g., flower-bulb vs light-bulb). It seems that one limitation of these homograph studies might be a deeper level question about the true nature of linguistic representation and, furthermore, textual representation. For example, if I were to see bulb with a context word of light, the activated representation will overlap very little with that of flower-bulb (apart from a shared representation of the motor movements rehearsed to produce the same sound). Although we represent these two things with the same arbitrary string of symbols (i.e., letters), it is unlikely there is any more than a trivial overlap between the two representations. Letters and words are used as symbols to represent something. If two different representations are activated sequentially, it does not seem very surprising that the effect is not as pronounced as just seeing the word twice or seeing the same word-cue pair twice. As such, I am not sure this particular challenge is as compelling. However, I feel the additional experimental evidence and the arguments were more than sufficient to illustrate the limitations of the “reminding” theory. Overall, I thought this was a great article and an excellent example of attempting to more precisely examine the mechanisms underlying an effect, rather than just examining when it works or doesn't.
This review article explained spacing effect in terms of various theoretical accounts. I assume that most of students are familiar with this spacing effect phenomena even though they have not ever studied working memory or long term memory because this phenomenon actually happens frequently when we study and get test. This review article seems nicely organize major theories so that I could get a clear concept of spacing effect, more than just feeling this effect in real life. Among proposed various theoretical arguments to account spacing effect here, deficient processing seems interesting since this idea generally indicates that if first processing of certain item is not deep enough to make firm memory (or memory cue) the quantity or quality of the item processing at second presentation would be reduced. It could be related with in memory formation how important first processing is , as much as repeated study, which is close to my experiences in real life. Of course, repeated learning is generally a key to maintain long-term memory as mentioned in this review. In terms of replication, I would like to see how powerful spacing effect is occurred in different type of learning, other than verbal learning, such as mathematics, musical training, or any kind. I guess I definitely could see similar spacing effect from those types of learning, so it might be helpful to strengthen the theoretical address of spacing effect.
It wasn’t really until this semester that I had become acquainted with the concept of “spacing effects” at all, so my understanding about the idea and its mechanisms was very minimal. With that having been said, as people who (presumably) spend considerable amount of time studying, there seems to have always been a sentiment set forth that cramming for things does not induce the same level of understanding as a more step-wise approach. But at the same time, increasing the lag time too much can also not have the most ideal effect either. I think that as students (and potential students), understanding spacing effects can ultimately help in the goal of optimizing learning. What seemed a bit counterintuitive in the article was their criticism of meta-analysis as exclusionary, yet in this very review uses a method that also requires the researchers to sift through the various materials and come to conclusions about them. With regards to replication— I attempted to have the theme in my mind, but have come up with little to say about it. A noteworthy aspect of the article was that they mentioned 7 different mechanisms that could account for the spacing effects. Perhaps the article’s purpose was to point out that the over 100-year history of recognizing spacing effects (and extensive research that his been done), there are still competing theories. Conceptually replicating seems to be an area in which this research would greatly benefit, in order to give credence to one mechanistic proposition over the others-- especially if looked at within different domains.
I enjoyed reading this article a lot. I had no idea that there were so many theoretical mechanisms behind the spacing effect. I was excited when the article started to talk about the possible mechanisms, but at times I did not feel like they explained the possible mechanisms very well. Though this mostly applied to the mechanisms mentioned early on. I felt like they mentioned them very briefly and moved on too quickly, at least for my understanding. It seemed like the author was very familiar with the topic area (as he should be), but there was a gap between his understanding and mine. If he had just made a few things more explicit in the introduction, then it would have made the rest of the paper easier for me to comprehend. I found myself having to keep referring back to earlier parts of the paper or looking up terminology in other places. This may just be my sick self failing to process information completely. It just felt like a lot of information crammed into a small amount of space.
I think that it is an interesting thought that it could be multiple mechanisms combined that account for the spacing effect. This makes me think of what Art Graesser was saying during his talk last week, where when we do research like this, we try to simplify things as much as possible in order to come up with one specific simple answer, when really the way that humans operate is complex. So sometimes the simplest explanation is not always the most accurate one, or the one that is most explanatory. I also like the article’s brief mention of individual differences in how people experience the spacing effect. I am not very familiar with this area of research (assuming that it exists, which I feel like it must), but I would be interested to see what individual differences influence the spacing effect.
This week's article by Maddox was somewhat unique compared to the others we have covered, in that it was a review article rather than a direct study and addressed a phenomenon that seems to have been replicated in a variety of ways over many studies. In terms of replication, the article seemed to address different instances of conceptual replication more than direct replication, even going as far as to mention that the spacing effect occurs across different animal species. Though not a specific highlight of the article, mention of this finding stuck out for me, as replication of cognitive phenomena across species would seem to be one of the strongest instances of evidence in replication, at least when it comes to lower-order processes.
In terms of the mechanisms underlying the spacing effect, the debate over whether one particular mechanism or dual mechanisms are accounting for the effect seemed to highlight an important issue in replication, namely, how to manage parsimony when developing models and attempting to replicate them. For instance, the authors mention that one explanation of the spacing effect highlights encoding variability and study-phase retrieval while another attempts to be more parsimonious by reducing the explanatory mechanism to a single phenomenon. In these cases, particularly when one explanation overlaps with or even incorporates another, it would seem difficult to determine which factors or processes are being effectively replicated. For example, findings from one study may provide support for a number of important underlying factors while another may find support for only one, but equal instances of the two scenarios makes it incredibly difficult to determine where a model has become parsimonious enough while still being capable of depicting the complexity of certain processes. This may then be even further muddled if some mechanisms operate with slight difference depending on context. This issue exists in a range of disciplines and across topics, and issues with replication tend to fuel a perpetual fire of debate over the underlying structure of certain constructs. This is the case for a number of constructs in Educational Psychology, such as achievement goal orientations, for which many debate a 2x2 framework while others espouse evidence of collapse in constructs.
One last thing that stuck out to me in the article concerning replication was the author's comments on meta-analyses. Meta-analyses would seem to be one of the most useful tools for getting a big picture view of replication in a given domain, and the article does mention a number of meta-analyses supporting theories of the spacing effect. However, as the article mentions, meta-analyses can be undermined by the inclusion criteria they implement. Though I had understood the need for inclusion criteria in conducting meta-analyses, what I had not considered outright was how researchers' underlying perspectives and theoretical understandings might influence how they set criteria and evaluate findings. This raises a unique issue in replication, as I would think that popular theories, regardless of how accurate or comprehensive they are, would dominate meta-analytic findings, while alternative explanations may fall to the wayside, with corresponding replication evidence going unseen.
This week we have a review paper on the spacing effect. I've read some spacing effect papers and heard Phil talk enough that I have a rough idea about it already.
The thing I was always wonder is what is the optimal number of times for something to be repeated and what is the optimal duration between them. I haven't been able to get a clear answer, and looking at Fig 1 and Fig 3, they seem to show two different patterns: Fig 1 increases up until the optimal point in which case it then decreases but Fig 3 shows that it reaches some optimal point then levels off. Fig 3 also shows that fewer events do better than more which seems really weird!!!
Another thing that I have thought of before when Phil was talking about the spacing effect with very short durations, is how is that different than cramming? Just like the section 'Lag by retention interval interaction' discusses, if the lag time is on the magnitude of 20 seconds, is that really any different than cramming?
That being said, my main concerns on what are the takeaways and how can I apply this, rather than what are the underlying mechanisms at play.
The Maddox article was a different sort of view on replication from what we have seen thus far, in my opinion. While many of the articles we have gone over and presentations we have seen seem to have primarily focused on the issues and controversies when findings are not replicated, this article beautifully seams together both studies that have and have not replicated and how this supports or does not support theories. In part, this is because this article is a review and not a study, but I still thoroughly enjoyed the set-up and transitioning throughout the paper.
ReplyDeleteI am curious how studies/articles were chosen to be included in this review, though there is no apparent reason presented in the paper to believe selection was heavily biased. To me, the reasoning behind how the theories presented were and were not supported was soundly defended.
The implications for the “real world” seem to be particularly relevant for this review, as this is a phenomenon that is seen in classrooms and outside of the laboratory as explained in the article. With a better understanding of the mechanisms behind spacing effects, things like workshops, training sessions, classroom materials, etc. can be better designed to take advantage of these mechanisms. As well, with a better understanding of theory can come better research design to examine this effect further.
This area this week’s article focuses on is very new to my knowledge. It contains a lot of information about the research that has been done on spacing effect. The author first pointed out the limitation of meta-analytic approach in the studies of spacing effect, which is “the selected search and inclusion criteria can limit the scope of the review”. Then, the author posited his solution to address this limitation by examining critical findings in spacing effect literature that have been absent in prior review and assessing past and present theoretical accounts based on the consistent critical findings. This approach of reviewing findings first, then evaluating the extent to which prior theories can accommodate the findings is very different than meta-analysis studies I have read so far. I am wondering whether this way of review can be replicated to other field of knowledge to address the limitations of a meta-analytic approach.
ReplyDeleteGenerally speaking, since long-term benefit is the goal, so a lot of studies reviewed in this article focus on how spacing effect influence final test performance. I am thinking whether the critical findings or different account can be consistent in different domain of learning in terms of conceptual replication. In addition, it seems that the combined encoding variability and study-phase retrieval mechanism can accommodate all consistent findings in the literature. The author posited that when encoding variability was too much, it may lead to a failure of study-phase retrieval, the question I have here is that whether the variability is too much or not is domain specific. In conclusion, for future replication, it will be interesting to examine spacing effect on different domains, different forms of learning tasks, and different level of learning tasks.
Coming into reading this article, I thought that I was fairly informed when it came to the spacing effect. As I read this article I learned that I was very wrong ;) As a review article, I felt that this paper did a fantastic job of covering all of the nuisances of the spacing effect and I particularly enjoyed how each section devoted time to tying back to previous sections to allow the reader to get a more clear picture as to how all these puzzle pieces may fit together. Each section built very nicely on each other—though I felt that some of the sections got a bit bogged down in jargon and made it a bit more difficult to understand how some of the different pieces fit together.
ReplyDeleteI don’t want to spoil to much before Alex and I lead discussion but I was particularly intrigued by how the breakdown of the specific mechanisms helped to better explain the spacing effect as a whole. I had not even considered incidental encoding in regard to the spacing effect before—though the phenomena itself is an interesting one. One other facet that I was also most intrigued by was the discussion of lag and how that effects encoding and retrieval. I mean, after I read the breakdown it made so much more sense that lag interval would have a significant effect on spacing and retrieval success % but I never even considered that as a factor before (I’ll save the rest of what I have to say for discussion).
As for replication, the spacing effect is interesting because it has already been replicated countless times for many, many years. I know the spacing effect is often seen in memory research but I wonder if the we may see the benefits of the spacing effect transfer to other domains—perhaps domains that require procedural memory and countless repetitions such as shooting a basketball. Would these types of memories qualify?
This week reading is interesting. My first impression about this article is the article is not a good example of replication. The author did nice job putting all the information and theories to support their argument but did not mention how replicated with the current research questions that the author looking for.
ReplyDeleteSpacing effect refers to the mnemonic benefit of spacing repeated study events across time compared to massing. The spacing effect was first reported over 100 years ago and refers to the long-term memory benefit produced by spaced study events compared to consecutive study events. The benefit of spaced study is further modulated by the specific interval (i.e. lag) that separates presentations of an item (i.e. the lag effect). Spacing effect research has been subject to numerous reviews which have been useful in developing and evaluating theoretical mechanisms proposed to account for the phenomenon. But at the end of the day I feel like how much people care about spacing effect right now in this technological advance world
The article also states two objectives to better evaluate the mechanisms proposed to underlie the spacing effect, 1) critical findings in the spacing effect literature that have been absent in prior reviews, 2) consideration of various instantiations of the encoding variability account which is beyond the basic assumptions. That provides as a means of assessing the viability of an encoding variability mechanism to explain the long-term memory (LTM) benefit of spaced stud. Methodology wise author did nice job fulfill this two objective very thoroughly.
The author did also mentioned that consistent findings from the extant spacing effect literature that had not previously been considered in concurrent fashion, and in turn this allowed for further specification of the theoretical mechanism underlying the spacing effect. The study-phase retrieval component of the underlying mechanism showed the individual and group differences will be necessary considerations when designing spaced study schedules. One of the limitations of the literature noted above suggest that there is still substantial opportunity to parameterise and examine the spacing effect.
Overall, I enjoyed this week reading !
I really enjoyed this week’s article. It was easy to read, and it was interesting to learn about all of the different aspects of the spacing effect that have been investigated. There were examples of both conceptual and exact replications, which I appreciated, considering the theme of the class. One thing that I thought was interesting was that the authors kind of talked down on meta-analyses, and how certain articles can be left out if they don’t support theories, but then I’m sure they had to pick and choose articles to include or exclude in their own review.
ReplyDeleteMy overall takeaway from this reading is that replication can really help theories develop. An experiment might have extremely interesting results that challenge previous research, but direct replications are needed to figure out whether those results are valid. It might be that researchers have uncovered the exception, not the rule. We know so much about the spacing effect because it has been known about and studied for over 100 years, but there are still unanswered questions, which this article really made me think about. As we learn more, research can be better translated to classroom settings, and learning can be optimized.
This article was a bit technical for me, so most of my commentary consists of questions and confusions.
ReplyDeleteIn the section on experimenter induced variability across sections, one condition described is a cue word change that biases the same meaning, e.g., flower-bulb and tulip-bulb. But these do not have the same meaning and, in general, it can be difficult to determine a word's semantic meaning. There are a few ways to describe or determine meaning, including extension, intension, use. On none of these are these two words the same (though in all cases they partially overlap). My concern is for other word pairs that may have been less similar but deemed the same.
I don't understand the hypothesis formulated by Vorkoeijen and Delaney, that "fast presentation rate would limit spacing between repetitions of massed items in covert rehearsal, whereas a slower presentation rate would allow for longer spacing intervals between covert repetitions of massed items." I suspect my issue is one of vernacular, though it might be my experimental ignorance.
Another point of vernacular: What is a rehearsal? Just the speaking aloud of the words in order, of one word? During presentation? After?
As someone who has used flashcards for language learning, the possibility that massed items produce a sense of knowing (but not necessarily actual knowing!) and thus leading to a decision not to rehearse massed items (at least as much as spaced ones) rings true. Of course, using this as an explanation assumes "control," and the appearance or sense of control is not necessarily control.
What is "qualitative" about the qualitative automatic deficient processing account?
Again, I enjoyed this week’s article. I personally have not read a lot of review articles and this one was great. I think it was very detailed and comprehensive, which is important of course. I am familiar with spacing effects but I would not consider myself to be extremely knowledgeable in this area, so I really appreciated the way the background information was presented. I do have to say, some parts were more difficult to “get through” than others, perhaps that’s just me.
ReplyDeleteI did get the sense that they were not exactly “for” doing a meta-analysis. After the recent meta-analysis, we discussed in class, I find it almost essential for researchers to explain what criteria was used, despite it being an actual meta-analysis or a review. This criterion allows us to have a better understanding of where the researcher is coming from, and I felt these details could have been included.
I’ve only ever seen spacing effects in research that involves memory, more specifically learning and education. I’m not sure where or why the thought of “would motivation play a role here” came into my mind, but I wonder how motivation could fit into this. Naturally, we think that yes, we reward someone after every time they study, they will continue to study and do better on the test. But do we know this for sure? Are there certain patterns in rewards or motivation that are more effective than others in terms of spacing effects? For example, if we were to have two groups of students with an upcoming exam, one who received extra credit in a course where they completed spaced online study actives and the other who did not, would we see a difference in motivation between the two groups? Or a difference in their performance overall?
This article was primarily interesting because of the research I do with Phil Pavlik and his ITS MoFaCTS. In short, MoFaCTS (Mobile Fact and Concept Training System) is an ITS that spaces concept/fact repetitions to optimize retention and remembering. I also thought I had a relatively solid understanding of the spacing effect, but now I am not so sure.
ReplyDeleteI wonder if this article delved too deeply? If we say that the spacing effect only works in very (VERY) specific circumstances, is there really an effect? I, for one, think there is a solid effect when items are spaced optimally....optimally being the key word there. It seems pretty straightforward to say that spaced practice inherently requires some sort of deeper remembering process than massed practice. That, in and of itself, seems beneficial to me. Although I am slightly biased, I am a big proponent of the spacing effect and its potential use in classroom/educational settings. Even if we are looking at spacing in a short-term memory way, spacing items for practice will inevitably keep that item in your short term memory "register" in a way that massed practice will not.
Looking forward to this discussion!
This week's article seems to be less directly related to replication than some of our previous readings in some senses. While it largely a review of the literature (certainly an enlightening one), it doesn't really fit into our ideas of directly or even conceptually replicating. What is does do is play an important role in leading the conversation.
ReplyDeleteThe important place for reviews such as these is that distill the conversation. This review was able to concentrate on consistent findings that already exist in the literature on spacing effects. Having a singular, well-written piece that relays the state of the field can be invaluable to researchers looking to replicate findings and advance the field. This particular piece helps to direct scarce resources into possibly enlightening investigations.
This was a great spacing effect paper. However, this review also did a fantastic job of demonstrating the Dunning-Kruger effect. That is, I began this paper thinking my peripheral exposure to the spacing effect literature through different classes and articles on memory was sufficient to help me understanding what it was. Now I realize I only know enough to fool myself into thinking I understand it at all. But, I digress.
ReplyDeleteIt was very interesting to see how the author used such fine-grained and nuanced findings to argue against a recent theory. I am not well-versed in the literature to know which explanation is preferred (or even if much thought is regularly given to the mechanisms underlying this effect), but I thought this paper did a good job illustrating the weaknesses of this “reminding” theory. However, I am curious about one of the findings raised as a limitation. The author discusses experimenter-introduced variability across repetitions in one section. Eventually, this leads into a discussion of several studies showing a reduced spacing effect for spaced items paired with different cues. Specifically, this is about the study where they showed a decreased effect in cases where homographs were paired with the same or different cues across conditions (e.g., flower-bulb vs light-bulb).
It seems that one limitation of these homograph studies might be a deeper level question about the true nature of linguistic representation and, furthermore, textual representation. For example, if I were to see bulb with a context word of light, the activated representation will overlap very little with that of flower-bulb (apart from a shared representation of the motor movements rehearsed to produce the same sound). Although we represent these two things with the same arbitrary string of symbols (i.e., letters), it is unlikely there is any more than a trivial overlap between the two representations. Letters and words are used as symbols to represent something. If two different representations are activated sequentially, it does not seem very surprising that the effect is not as pronounced as just seeing the word twice or seeing the same word-cue pair twice. As such, I am not sure this particular challenge is as compelling.
However, I feel the additional experimental evidence and the arguments were more than sufficient to illustrate the limitations of the “reminding” theory. Overall, I thought this was a great article and an excellent example of attempting to more precisely examine the mechanisms underlying an effect, rather than just examining when it works or doesn't.
This review article explained spacing effect in terms of various theoretical accounts. I assume that most of students are familiar with this spacing effect phenomena even though they have not ever studied working memory or long term memory because this phenomenon actually happens frequently when we study and get test. This review article seems nicely organize major theories so that I could get a clear concept of spacing effect, more than just feeling this effect in real life.
ReplyDeleteAmong proposed various theoretical arguments to account spacing effect here, deficient processing seems interesting since this idea generally indicates that if first processing of certain item is not deep enough to make firm memory (or memory cue) the quantity or quality of the item processing at second presentation would be reduced. It could be related with in memory formation how important first processing is , as much as repeated study, which is close to my experiences in real life. Of course, repeated learning is generally a key to maintain long-term memory as mentioned in this review.
In terms of replication, I would like to see how powerful spacing effect is occurred in different type of learning, other than verbal learning, such as mathematics, musical training, or any kind. I guess I definitely could see similar spacing effect from those types of learning, so it might be helpful to strengthen the theoretical address of spacing effect.
It wasn’t really until this semester that I had become acquainted with the concept of “spacing effects” at all, so my understanding about the idea and its mechanisms was very minimal. With that having been said, as people who (presumably) spend considerable amount of time studying, there seems to have always been a sentiment set forth that cramming for things does not induce the same level of understanding as a more step-wise approach. But at the same time, increasing the lag time too much can also not have the most ideal effect either. I think that as students (and potential students), understanding spacing effects can ultimately help in the goal of optimizing learning.
ReplyDeleteWhat seemed a bit counterintuitive in the article was their criticism of meta-analysis as exclusionary, yet in this very review uses a method that also requires the researchers to sift through the various materials and come to conclusions about them. With regards to replication— I attempted to have the theme in my mind, but have come up with little to say about it. A noteworthy aspect of the article was that they mentioned 7 different mechanisms that could account for the spacing effects. Perhaps the article’s purpose was to point out that the over 100-year history of recognizing spacing effects (and extensive research that his been done), there are still competing theories. Conceptually replicating seems to be an area in which this research would greatly benefit, in order to give credence to one mechanistic proposition over the others-- especially if looked at within different domains.
I enjoyed reading this article a lot. I had no idea that there were so many theoretical mechanisms behind the spacing effect. I was excited when the article started to talk about the possible mechanisms, but at times I did not feel like they explained the possible mechanisms very well. Though this mostly applied to the mechanisms mentioned early on. I felt like they mentioned them very briefly and moved on too quickly, at least for my understanding. It seemed like the author was very familiar with the topic area (as he should be), but there was a gap between his understanding and mine. If he had just made a few things more explicit in the introduction, then it would have made the rest of the paper easier for me to comprehend. I found myself having to keep referring back to earlier parts of the paper or looking up terminology in other places. This may just be my sick self failing to process information completely. It just felt like a lot of information crammed into a small amount of space.
ReplyDeleteI think that it is an interesting thought that it could be multiple mechanisms combined that account for the spacing effect. This makes me think of what Art Graesser was saying during his talk last week, where when we do research like this, we try to simplify things as much as possible in order to come up with one specific simple answer, when really the way that humans operate is complex. So sometimes the simplest explanation is not always the most accurate one, or the one that is most explanatory. I also like the article’s brief mention of individual differences in how people experience the spacing effect. I am not very familiar with this area of research (assuming that it exists, which I feel like it must), but I would be interested to see what individual differences influence the spacing effect.
This week's article by Maddox was somewhat unique compared to the others we have covered, in that it was a review article rather than a direct study and addressed a phenomenon that seems to have been replicated in a variety of ways over many studies. In terms of replication, the article seemed to address different instances of conceptual replication more than direct replication, even going as far as to mention that the spacing effect occurs across different animal species. Though not a specific highlight of the article, mention of this finding stuck out for me, as replication of cognitive phenomena across species would seem to be one of the strongest instances of evidence in replication, at least when it comes to lower-order processes.
ReplyDeleteIn terms of the mechanisms underlying the spacing effect, the debate over whether one particular mechanism or dual mechanisms are accounting for the effect seemed to highlight an important issue in replication, namely, how to manage parsimony when developing models and attempting to replicate them. For instance, the authors mention that one explanation of the spacing effect highlights encoding variability and study-phase retrieval while another attempts to be more parsimonious by reducing the explanatory mechanism to a single phenomenon. In these cases, particularly when one explanation overlaps with or even incorporates another, it would seem difficult to determine which factors or processes are being effectively replicated. For example, findings from one study may provide support for a number of important underlying factors while another may find support for only one, but equal instances of the two scenarios makes it incredibly difficult to determine where a model has become parsimonious enough while still being capable of depicting the complexity of certain processes. This may then be even further muddled if some mechanisms operate with slight difference depending on context. This issue exists in a range of disciplines and across topics, and issues with replication tend to fuel a perpetual fire of debate over the underlying structure of certain constructs. This is the case for a number of constructs in Educational Psychology, such as achievement goal orientations, for which many debate a 2x2 framework while others espouse evidence of collapse in constructs.
One last thing that stuck out to me in the article concerning replication was the author's comments on meta-analyses. Meta-analyses would seem to be one of the most useful tools for getting a big picture view of replication in a given domain, and the article does mention a number of meta-analyses supporting theories of the spacing effect. However, as the article mentions, meta-analyses can be undermined by the inclusion criteria they implement. Though I had understood the need for inclusion criteria in conducting meta-analyses, what I had not considered outright was how researchers' underlying perspectives and theoretical understandings might influence how they set criteria and evaluate findings. This raises a unique issue in replication, as I would think that popular theories, regardless of how accurate or comprehensive they are, would dominate meta-analytic findings, while alternative explanations may fall to the wayside, with corresponding replication evidence going unseen.
This week we have a review paper on the spacing effect. I've read some spacing effect papers and heard Phil talk enough that I have a rough idea about it already.
ReplyDeleteThe thing I was always wonder is what is the optimal number of times for something to be repeated and what is the optimal duration between them. I haven't been able to get a clear answer, and looking at Fig 1 and Fig 3, they seem to show two different patterns: Fig 1 increases up until the optimal point in which case it then decreases but Fig 3 shows that it reaches some optimal point then levels off. Fig 3 also shows that fewer events do better than more which seems really weird!!!
Another thing that I have thought of before when Phil was talking about the spacing effect with very short durations, is how is that different than cramming? Just like the section 'Lag by retention interval interaction' discusses, if the lag time is on the magnitude of 20 seconds, is that really any different than cramming?
That being said, my main concerns on what are the takeaways and how can I apply this, rather than what are the underlying mechanisms at play.