Practice As Research
Practice As Research
Investigating cognitive processes in language learning: The use of eye tracking and related ethical considerations
In the last decade, the field of second language acquisition has witnessed an increase in the number of studies using eye-tracking to examine the cognitive processes involved in language learning. Eye-tracking allows researchers to record learners’ eye movements while completing a task on a computer screen and provides a very rich record of online processing behaviour. It is increasingly used in the field as a measure of cognitive effort. In this presentation I will provide an introduction to the eye-tracking technique, as well as a brief overview of some of its applications in language learning research, with a particular focus on vocabulary learning. The last part of the presentation will discuss the ethical considerations in this type of research.
Dr Ana Pellicer-Sánchez is Associate Professor of Applied Linguistics and TESOL at the IOE, UCL’s Faculty of Education and Society, UK. Her research focuses on the teaching and learning of vocabulary in a second/foreign language. Her recent research has used eye tracking to examine cognitive processes involved in vocabulary learning, with a particular focus on learning from reading. She is co-author of An Introduction to Eye-tracking: A Guide for Applied Linguistics Research (CUP) and co-editor of Understanding Formulaic Language (Routledge).
Hello, and welcome to the practice as Research and Network Seminar series. Um! It's really exciting to have everybody here. Um! It's been quite a few months since we've had the last of this installment over the summer break, and then, obviously, Covid related um illness, one. We had to to postpone one of our talks as well. But um! Hopefully, we're back um into this ring of things now, and we can do them regularly again. Um, I must say that I have quite missed them. So I'm really really excited to have them back, and it's really even more exciting for me. Um! Today we have somebody here who is our guest presenter, and who's actually from my own um department at work. So this is, you know, doubly exciting, because it's always nice to hear other people talk from different different disciplines. But um, you know, when you actually realize that actually you've got some really interesting stuff going on in in the very place where you work, and you don't usually get time to talk to people's research. This is a nice way of actually catching up with people that friends. So I do enjoy that as well. Today we have got with us, Dr. Anna, and who is going to be talking about? Um what it means to to be ethical when you're researching cognitive processes and language learning um, and specifically, she's going to be talking about eye tracking um, and i'm quite excited about that, because i'm very conscious when i'm sitting in front of the camera like this. I'm looking into the camera. Not my screen. Um. So you know the eye tracking thing is kind of at the back of my mind as well. So that's that's something that's really exciting and to kind of introduce, and um i'm very briefly. Um. Dr. Anna Blythe S. And Jess is um associate Professor of of Applied Linguistics and tees all at the Institute of education. Uco's Faculty of Education and Society in the Uk. Two. Her research focuses on the teaching and learning of vocabulary in a second or a foreign language. Um, and her recent research has used eye tracking to examine the cognitive processes involved in vocabulary learning with a particular focus on learning from reading one hundred. Um anna is a co-author of an introduction to eye tracking a guide for applied linguistics research and co-editor of Understanding for Malay language, um published in outlets one hundred and fifty. So, without any further I do. I'm going to hand over to Anna, and what i'm going to do is i'm going to stop sharing my screen first of all, and then Anna. And if you could just, you know, do your presentation. First, I will keep my video on so that you can see me so you're not talking to it. I have no worries, but I will mute myself, so there is no um, you know no, no echoing or anything. Um! So when when i'm when you're talking, i'm going to mute myself, and then after that Um, we're going to sort of have a little bit of a discussion, and I always take this opportunity to kind of ask questions first. This is my prerogative as the person hosting it. So i'm always getting the first question, which is really nice. Um! But everybody else in the room. Please do feel free to raise your hand. Virtual, real, to unmute yourself later. Um, but also to raise any questions that you may have in the chat box, which I will keep an eye on as well,
Ana Pellicer-Sanchez:so you please and I, if you would like to share your screen. It worked before us. So that's how it does work here. Uh, can you see the slides? Just the slide. Yes, perfect. Excellent. Okay. So i'll get it started. Well, thank you First of all. Thank you very much uh Nicole for the invitation. It really is a pleasure for me also. Uh, it's always a good opportunity to talk about our research and what we do. But this has a slightly different focus, and it's a slightly different uh audience. So it's always good uh, cause i'm sure I'll I'll, uh, you know, will be an interesting uh discussion, and I will learn from these as well. So it's quite a yeah. It's quite exciting for me to be to be here today, as Nicole was saying. Uh, What I will do is um about. I'm going to be talking about how we use eye tracking to explore cognitive processes in language learning unrelated ethical uh consideration. So um what I thought I would do i'm not sure how familiar everybody is with eye tracking, so I would provide a very brief overview of what it is. Uh, what will the technique involves, and also and um provide a brief overview of some of the key applications in language, learning those things that I've been working on for the last a few years uh focusing on vocabulary learning one of the topics that I've explored using um I tracking, and then at the end of listed a few of the main ethical considerations that arise from this research. I'm sure that you can think of more. So So we leave that for the discussion. But of yeah, I've included some of those main ethical considerations that arise from this type of research. So uh, I, tracking is basically uh, uh, for those of you who are not familiar with these is basically a device that we use to measure learners or participants eye movements while they're processing any sort of a stimuli on a computer screen. So it could be a text. It could be an image. It could be a um, a a movie. So anything that we display on a computer screen. Then with these, with the I tracked you guys, we master participants eye movements while they're processing these different types of stimuli. There are many different types of um equipment and different types of systems that we could use, that I'll talk about that at the end, because, of course, considerations are different, depending on the one that you are using. Um, So it provides a uh, a very rich record of processing behavior. So where participant participants and learners in in my case are actually looking at, and for how long. So, from the very early days it was considered as an opportunity to look into the mind of the subject. And this is key why we're uh, We've started to use it in language, learning, research as well. So, as I was saying, it measures eye movements, and in particular it measures three different types of eye movements, saccades, which are the very fast movements of the eyes. So when we move from one point to another, when we're reading a text, or when we're uh, you know, looking at a picture or watching a movie uh. So the movements, the rapid movements of the I would be uh the calls account fixation. So basically this stops the times where we stop and fixate on a particular part of the text, or uh image, or whatever uh the stimulus uh is, and regressions this is particularly for reading. Uh, these are movements back in a text. So we are reading, and then we move back to. We read all the parts of the text, and also an example, so that it's uh clear. So imagine that this is a sentence that we are presenting uh people uh with, and and we want to know how they are processing these types of uh, this particular sentence. So the grace uh uh circles here on the screen are basically fixations. So number one to six are the stops, the points where uh our eyes is top to process information, and these are the points where um information is actually processed. There's not that much information process during the account. So it's mainly those fixations that those are stops of our eyes what we process information. So the arrows a um to e would be um the Us. Accounts. And then uh, at the end we have an example of our regression. So we're going to start reading this sentence. The woman saw the man with binoculars, and then we go back again, and we read Man Ah! And the end of binoculars again. So these movements back would be regressions. But I won't go too much into the tech. You know the details of the measures and all that, just for us to have an idea of what types of eye movements we measure with these um equipment. So the important thing is the assumptions behind these. So one of the main assumptions. Uh behind the the the study of eye movements is that the the amount of time that we spend fixating an item um stopping to to process a particular item. When I say item, it could be, you know, words, or a part of an image or or dynamic uh image, it reflects the cognitive effort required to process it, and the what we are fixating is what is being considered. So when we fixate on a particular word or part of uh, of a visual display that, uh, reflects the cognitive effort that is involved in processing, and that this is actually what we are considering right. So it has many advantages. It can be done without a secondary task. And this is important for you. You know experiments that actually explore at things like, you know, processing time and reading Um, because you don't need to ask participants to do anything other than just sit back and read whatever we present on the screen or watch this movie. They don't necessarily need to press the bottom. Or do you know, we record any sort of measure um any sort of response, as as as of their techniques, would it? As I was saying earlier, it provides a very rich record of processing behavior, so we can know exactly where participants are looking at and for how long. Um, So I talking data is normally reported in terms of number of fixations. The number of how many times people is stopped in that particular area that you're interested in, and the duration of those those fixations. So for how long? Um, and whether that particular area was fixated or was skipped. Um and data is often also reported in terms of early and late message. There are loads of different eye tracking measures that i'm not going to get into. But it just some of those reflect more automatic processing and all there's more control is strategic processing, so that just for for for us to know. So there are challenges when we say in the last. So I track it was obviously very much used. It was a common technique in cognitive psychology and and and psycho linguistics for decades. It's been really in the last ten years of so that it's It's really uh been started to be used in in language, learning, and and second language acquisition, research, and there has been a huge increase in the number of studies using it, because obviously of the benefits, and how useful it is to explore Ah! Underlying cognitive processes there are important challenges as well that we need to consider um. So, especially when we use this technique as I've done as we are doing in in the field to explore language learning processes. We have to be careful about how representative it is some real learning situations. I mean, remember that in the majority of cases we are Um, I mean, data is collected in a lab in an ice-racking lab with the students one by one I mean. It's very difficult to make real classroom a and real language learning situations. I mean, they're very few exceptions around the world of you know, labs where they have. I don't know ten or fifteen different equipment, but they're very expensive, so that's very rare. So In the majority of cases we have one or two uh I trackers in the lab, so that's something to to consider. Um. It's important to consider as well. But in these kinds of in this kind of more applied language, learning, context, um I tracking, and and this the the the examination of eye movements cannot really tell us everything about cognitive processing. So we'll be able to say, Okay, learners spent these amount of time on this item in comparison to these older they they seem to pay more attention to this part or the other, but we very often it's not possible to know why exactly. They were spending. You know more attention on a particular part of the stimulus. So that's something to um to consider. And very importantly, eye movements are very precise um measurements that they're affected by many, many different factors. So the actual appropriate design of I tracking is that is key, and it really the design is related to how you know the the the claims that we can actually make uh based on the data. So i'll show an example so that you you You see what I mean. So this is a um, an image from a from an ad um. This is a heat map, so it's just a visual representation of eye movements. The red areas are basically the areas that received most attention. So if we look at these, what can we say about the processing of these advertisement. Well, we can say that this participants seem to spend more time reading the title, the first part of the text, and definitely the the face of the baby. If we look at another representation where we see the the the order of the stops of the fixations, we could say that they first um look at um at the text. Well, that's probably a London fixation, but anyway, and then the baby, and then finally the title. But we can't really say um much about the processing of a specific words in the text, mainly because um, the text is too small for us to be able to make claims about that. So this is just to show how important the design of the materials on the stimuli in this type of eye tracking um experiment. So, when possible. It's good to modify materials and and and create them ourselves to make sure that the data can answer our research questions, and if we use authentic materials we will, There will be a range of variables that we will have to control, for in the analysis, or you know, that might just limit the the the claims that can be made. So I think this is really important when we're thinking about applications of eye tracking to different areas. So, as I was saying. It has received recently a a huge amount of attention from language, learning, and second language acquisition researchers. It has been applied and used in many different topics. Um of particularly used it it to explore the process of learning from written inputs. So learning from reading, particularly focusing on vocabulary learning in the majority of the studies. And also I've looked at a learning from multi model exposure. So when people have access to, you know text, but also images audio input um dynamic images. So how how attention is actually played in those cases? But i'm just gonna focus on learning from uh from reading today. And so you a couple of examples of of basically what we found from these uh from these research. So focusing on uh vocabulary learning from reading Um, when we talk about. Uh the different learning approaches and vocabulary learning approaches. We usually make this distinction between um A more intentional learning and incidental learning approaches so intentional learning when there is a clear intention to learn vocabulary and a clear um, a focus on learning, a set of words like when learners complete vocabulary activities, there is a clear goal to learn vocabulary, and more incidental approaches at those uh well, and the learning happens as a sort of a consequence of the of the of the reading activity as sorry of the of the communicative activity. So the focus is on meaning, not necessarily on phone. So um, when we talk about incidental learning. We're basically referring to learning, from reading, reading, while listening, listening or viewing. So there are lots of research showing that uh, just when learners read for pleasure, or watch a movie or listen to a podcast. They also improve their vocabulary knowledge, and they can learn new words from that, even if their focus is really just on the meaning and the content of um of whatever it is that they're exposed to. So loads of research has uh provided evidence for these. I'm really focusing today on learning from reading. So the main myth uh kind of design of this is that is the traditional design of uh uh studies looking at vocabulary learning from reading is that what we have, uh, you know, a pre test of reading activity, and then a post test. So in the pre-test, we measure knowledge of a set of words that are going to appear later in a reading text, and then, after the reading we measure knowledge again to see how much they've learned. And this design has been great, and this kind of um research design has allowed to to really um find out more about what learners get from reading what they remember after the reading activity, what they can remember about those new words that they um were exposed to one hundred and fifty. But it doesn't really tell us much about what happens when learners are actually reading the text, and they find words they don't know. I mean we. We, of course, hypothesize that, you know, in the majority of cases and learn is, could keep the words, or they could try to guess from context. But we really don't know, and I, tracking really provides a great means to actually find out what happens as they are reading the text. So this is certainly uh um something that has been uh um explored in in in vocabulary learning, research in the first language context. So looking at first language reading um a studies of found in general, as we would expect that when first language readers are reading a text, and they find words they don't know they would spend more time reading. Ah, those new words when compared to words that they know very well, and that reading times decreased with exposure. So that means the more times you see another word the faster your reading becomes, and that makes sense. That means you are, you know, building your familiarity with that word and learning uh that new word in the second language context. And then, as I was saying, the last eight years or so, they've been loads of studies, and that have trying to, you know, use a I tracking to to explore uh vocabulary Learning from reading, we found a similar pattern. So studies tend to show longer reading times for a known vocabulary. So when L. Two learners second language learners are reading a text, and they find words they don't know they tend to spend more time processing. They they spend more attention. Uh they pay more attention. Sorry on those words that they don't know um reading times decreased with more exposure, so the more time they read it, the um, the um, the faster the reading and more fluent the reading becomes. Uh, we found similar patterns for single words, and also formulating sequences like like allocations. So items beyond at the single word and um also importantly for language learning researchers, some studies, not all, some studies. I found a connection between the amount of attention that we pay uh that learn is paid to words in reading, and how well those words are learned. And that's key, because that's one of the main at theories of of of a language learning this idea of noticing, and i'll i'll I'll get to that in a minute. So So far, so good, and I, tracking has really allowed us to to learn a lot about vocabulary learning. But, um! We also know that, uh incidental vocabulary learning from reading, and any other source really really depends, as I was saying, on the involvement on the degree of involvement in processing. So the degree of attention learners need to pay attention to those words, otherwise they need to notice those novel words in the input, otherwise uh charges are that they will not learn those words. And this is it's really related, as I was saying to um um important theories in second language acquisition, like noticing hypotheses, or in depth of processing hypothesis, So an important question for a vocabulary, researchers and sla research is to to find ways to promote that, and make sure the learners actually see and pay attention, and notice those words that they don't uh that they don't know
Ana Pellicer-Sanchez: so um many different methods. Uh:I've been uh explored in the in the literature. One of them uh one of those that has received the most attention, I think, is this idea of premium modification of input So um things like
Ana Pellicer-Sanchez:um bolding underlining as you have here, the example here with highlighting, so all these techniques seem to support the noticing of the um of those items, and then leads to higher um uh games, and we have a studies both in vocabulary and grammar learning as well, and the couple of studies that have done this with eye tracking, has also shown that all these techniques of bolding, underlining actually lead to more attention. So they're good attention drawing uh devices. But one of these methods that I was really interested in, and I've done a few a a studies on this area. This is what I wanted to uh to to to you. Uh, in a very brief uh uh, four months. Um is instruction intervention before reading those of you. I'm not sure whether uh um um! Some of you are involved in in in language teaching, or well you might have experience as as language learners. This is a very, very common Uh um um activity in the language classroom, where we teach a set of words before we actually find them in a text. So if we, if there is a set of words that we find are going to be very difficult, for learn is, we teach them first, that relevant vocabulary, and then we ask them to read a text containing those um, those words. So the vocabulary initially learned through that explicit intentional learning, and then it it is consolidated. Uh in in reading, so it supports, uh lexical inferencing from reading as well. That's what research is upset, and there is some empirical evidence showing that it actually works for vocabulary learning, and it makes sense. That's what we would expect. But now there have been. Research is upset different things about the effect of these in structural intervention before reading uh on on the actual processing. So some researchers have said that it might increase the sales of the words and make learners pay more attention. But also the researchers have said that it might discourage guessing a strategy. So, since they already know what they mean, they might not pay that much attention to the words, and that might affect the learning process. And this was something that I was really interested in and thought, Well, I, tracking can help here, and we can really see whether teaching words before reading actually impacts, how learners deal with those words while they are reading the text, and that's precisely what I did uh, together with uh colleagues uh um uh Kathy Conkling and Lara, we kind of in in a couple of studies that we recently uh conducted, basically we compared, i'm not gonna go into the details of the study. And you can ask later if you, if you, if you have questions. But basically we compared uh pre reading instruction. So participants who actually receive teaching of a set of act of uh vocabulary uh items, a set of words before reading them in in in a text with participants who only we're only exposed to those words in a text, so they received no instruction whatsoever, and we had a control condition where they read a text with a real words uh and familiar words that they knew very well. So this is, for example, one page of the text that they read it contain some pseudo words, and we'll talk about that at the end. So they have to read a short story story with these Uh um! Unfamiliar items repeated eight times throughout the story, and we recorded the eye movements in the different conditions that I mentioned. So after the reading that we measure how much they had learned about the words, and then We also looked at the oly movement and data. And basically what we found was that in terms of vocabulary tests pre-reading instruction led to the highest scores and that's expected right. They receive instruction. And then they read the text. So in terms of the actual uh vocabulary, not into it, was the best group uh compared to reading only now in terms of eye movements, we have here a summary, And basically what we find was that the very first time that learn is so. Those words There was no difference whatsoever between the pre reading, instruction and the reading all the conditions, so that means it didn't matter if they had been taught the items or not, they actually spent a very similar amount of attention, which was against what we expected. Um! And of course, way longer than the control item, so the word that they knew very well, and towards the end what we find is a an advantage of the pre reading instruction. So there were no differences at the beginning. But once they had seen those words eight times, the participants in the pre reading instruction condition were actually able to read those new words in a very similar way to control items. So it's towards that they knew very well, So that means that the reading was much more uh kind of fluent, and when compared to reading only, so that points to an advantage of a pre reading instruction for um um for the processing of this new vocabulary. So it is really um. With these I just wanted to show the kind of things that I tracking can. Can Can Um, um! Well allow researchers to actually um know about the process in this particular case the process of vocabulary learning from reading. So it's been great so far in really allowing us to see not only what happens after, and you know what what kind of knowledge they remember from the words, but actually how that learning happens, and the amount of effort, the amount of attention that is put in to dealing with these unknown vocabulary, which we know is something that learners actually deal with a lot. So still a lot to do. But I think we've accumulated quite interesting evidence for the last few years. Now, ethical considerations. The first one has to do with the equipment. So the first thing that we consider with this type of experiment is any potential, you know, equipment related risks. I've put different pictures here that the first two where you actually have to wear something as a sort of weird helmet. This is the I track that I use when I started uh uh learning and and using eye tracking, and that was no good at all in the sense that it was heavy. And it was it it got warm? Uh, um uh, with time. So we had to be very careful about that. We had to be very clear about that in in information uh sheets for participants we have to be clear, uh very careful with the duration of the experiment, because, of course, people couldn't be wearing that for four hours. Uh, I mean, it would have been ethical to ask them to do that. So we had to be very careful about the duration of the experiment and be clear about that. Uh, you know, in the information sheet, Another important thing is um that usually I mean there with I tracking. What happens is that we have a camera and infrared light that captures your pupil, and we have to be clear There are no known uh uh risks for vision. Uh, but usually participants need to know uh these Now we are have, Uh, we are lucky in that. The systems that we use are much less uh inclusive. So there you it depends on which system you use. But uh, at least the one that I use is head-free. So you don't have to put anything in your head. Sometimes we ask people to put their chin in a chin, rest or forehand uh rest, so to control for movement. Uh! And again, this is something important, because if we do that, we have to consider how long the experiment is, for how long can people actually see it still without moving. So it tends to be relatively sort of experiment, or with uh breaks, so that people can actually move. And, um, we have to be very clear about that that they can. They can stop at any point, and they can, if they decide, is too much. Uh they can. They can leave. Basically and and what I tend to do as well is to include a practice session, a sorts of practice session at the beginning, so that they get an idea of what it is, and how it still they need to be, and and things like that. So that's the first um a kind of ethical consideration that, uh, that, I think you know, we need to consider. Another issue related to the equipment is the um, the issue, and this is important for for a participants consent is the um, the calibration I won't get into that. But getting a good calibration of the equipment is key to get good quality data. So the first thing that we do in this type of experiments to set up the equipment is to do a calibration, usually a nine point calibration where you would see something like when you see in the screen, and you have to follow those dots is basically about the equipment records the position of your eyes and can provide good data. Now. Sometimes it doesn't work, and sometimes you try again and again, and it has to do with the size of the pupil of the participant, or in the contrast it it really sometimes there are no explanations, but very often it happens, but not very often, but sometimes mit ctl, and and sometimes you cannot proceed with the experiment. If you don't get good caloration and good validation, two hundred and fifty, and an important question, I think, for us as we searches is, what what do we do if calibration doesn't work? Um. First of all their implications? Very often in these experiments, participants receive a compensation for their participation. What do you do, then? Do you steal that? Give them that compensation, or you don't? Um? But then you have to be very clear about that. In you know, ethics, applications, and information sheet for participants. They need to know if for some reason it doesn't work, are they going to receive compensation or only part of it? Uh, I think we have to be super clear about the the the the likelihood that this might not work something that I also I've dealt with is managing a students. These appointments. Sometimes they come, and they really want to participate in this study. And suddenly you tell them Oh, no, it doesn't work. You can't do it. Um. So something that I've done in the past is to have plan B and have another, you know, maybe a pe of paper version of some of the tasks, so that they actually do something, and they don't feel like they Haven't done anything, or just run the study if you can, even if you know the data is not going to be good at all. So these are important decisions that you have to make beforehand, and and have a very clear protocol of what is it that you're going to do in cases where it doesn't work. Um. Another important consideration has to do with particularly this topic, not with all eye tracking studies, but with this topic that I've talked about today is the task instruction. So in If we are exploring incidental learning conditions, participants cannot be aware of the specific aim, because if they know that this is studies about vocabulary learning uh that the data is like you can just throw it away, because uh it's not an incident, a learning condition mit ctl. And So this is a super important ethical consideration. Uh, when you know we are designing a studies like these, Usually what what I've done in the past is of course, explaining the ethics uh application form two uh why it's not possible to um to do this uh to to really reveal this specific aim at the beginning because of the nature of the research. Usually we would tell participants that we we will give them a more sort of general aim like this about language learning, or this is about reading comprehension. Um a, and that's what we would tell them in the information uh sheet. But then, of course, as soon as they finish their full disclosure of what they, What the specific aim of the study was I mean. Sometimes they've already realized, because they've done a few vocabulary tests, and then at the end, I think you know, they might say. Oh, I thought this study, you know, really had a focus on vocabulary. Some of them might say that, but it's important that we explain that to them at the end. But it's super crucial uh, in order to do research in this area to actually not disclose the specific aim at the beginning, because otherwise what you're just basically researching a different topic. Uh. So that's another very important uh distinction, and the last uh, the last consideration. Sorry, and the last point uh I I i'm sure you were thinking about that when I mentioned at this earlier, very often in these research, and and this is not just for eye tracking, but it's for any vocabulary related to vocabulary learning uh research. It's. A very common methodological decision is the use of pseudo-words. So words that actually do not exist in the language. They are invented. Words that look like real words, and this is very common in L one and out to vocabulary research, mainly because it has the advantage that um, you know it's a good way of controlling for previous knowledge. If you use real words, then you really need to make sure that you control for any sort of previous knowledge that participants had before this study. Uh. So that's a very common decision. But of course there is an important ethical consideration here. Uh, and I think it's particularly the case when we are doing classroom basis studies. So you're asking them to spend one hour of their time doing something to learn words that Don't actually exist in the language. So I think that's a very important ethical consideration. Um. And
Ana Pellicer-Sanchez: yeah, different researchers have different opinions about these. My um uh, my take on these, or what I've decided to do is really, you know, I don't do that when we are really uh um doing a studies in classroom um context. So when we actually using their language, learning time Uh:so we tried, you know, we would use real words and account for previous knowledge in in in other ways. So I tend to use pseudo-words only when these are really
Ana Pellicer-Sanchez:uh basis studies. They're optional. They come, uh, you know, at their own uh you know, outside their normal language, learning time. It's an extra activity, and I really try to emphasize what are the benefits. So you you know you won't be able to use those words that you've just led. But you know you've been exposed to these and this other thing, and you know in the text, You're also learning about all this. So I think it's important that we also think about what are the benefits there needs to be still a benefit uh um from from the experiment. And I think on this study, I think that's that's really really important. So these are some, of course, not all, but some of the ethical considerations of this type of um of research. And And yeah, And I think there should be a time for questions and time for comments. Now,
Nicole Brown: Absolutely. Thank you very much. I'm: I'm giving you a round of That's That's okay. Um, Thank you very much, Anna. It's great. It was really exciting. And like I said i'm, you know I'm:I'm quite, quite pleased to be the host of this, because you basically get means that I get to ask the first kind of question. Um! And and there are a number of things that you've mentioned. Um where I thought, Oh, this this is interesting. We need to talk about, And this is interesting. We need to talk about.
Nicole Brown:So. Um, i'll see whether I can. How many of my questions I can actually get through. Um! Just a quick reminder. Anyone who's here. Please feel free to pitch in with any questions or comments you may have. Um! Just make yourself known by either switching them video on, unmuting yourself, raising your virtual hand or by putting a um a note to me in the chat box. Um. So one of the things that i'm going to ask straight away. Um, I mean, first of all you did say about you know the sort of the random nonsensical words that people and um, I mean, would it be possible nowadays to say that actually we can still do that in the classroom, in your personal view, not not in the but in your personal view. Um. Given that in the Uk. And there is now this push in the in the early years where the children have to learn to read nonsensical words, so I mean it's the nonsensical words. I totally agree with you. It's it's quite quite a demand to say to a child? Well, you know, you know, learning ten words, and actually they don't mean anything, but because that's part of the everyday experience. Now that make it more acceptable in a way. Yeah, absolutely. And I think I mean these. These nonsensical words are also very much used in tests where um an activities to train pronunciation,
Ana Pellicer-Sanchez:and and you know, phonics and things like that. So it's actually something that they will be um, you know we'll be saying also by doing that. Yes, you won't be able to use those words ever. But actually, you are improving your your ability to, you know. Connect a a phony and graph teams uh, and that in theory that is supporting reading, development and language, learning, development. So yes, I I totally agree. I think that's one way of of saying, Well, you know, this is not vocabulary that you're going to use, but it could um! It could really be a uh something that that supports your learning. Uh in general.
Nicole Brown:Yes, no, absolutely. I mean I I just find it interesting. Um, you know, in terms of the obviously the um, the the ethics, and and i'm going to ask you, You know how easy or how difficult is it um to actually get this kind of work approved, irrespective now of whether you're using the nonsensical worse in a crack classroom environment. But if you're not actually telling your participants beforehand what exactly it is that you're looking for. I can imagine that ethics approval. Boards
Ana Pellicer-Sanchez:Yup may not be terribly excited about it. Yes, it is. That's why I said, It's one of the first things you have to be really really clear about. What is it that you're telling them? And why citing research? Uh uh saying? Well, you absolutely can't do that. Because then basically you're not doing research in this area, and you really need to um to to to to be very, very specific about what is it that you're going to tell them? Uh, you know at how they might feel about it when when they find out uh uh in the end? So yeah, this is a section in the ethics application form that I tend to explain very clearly, and and, you know, try to provide enough details. It It Hasn't, been a problem in the past as long as you explain very clearly why and what it what goes in the information sheet and what you're gonna tell them in the end and all that. It Hasn't been too problematic, but certainly something to to address in a lot of detail.
Nicole Brown:I think I mean, I mean, you know. Obviously um in my role at at at our Institute um in terms of, you know, looking after the ethics um for for staff members. Um, you know I I can definitely say that we are generally taking the the the view that it's not our job to stop research at all our job is not to stop any kind of research from happening. Our job is just to make sure that the ethics have been thought through, and that you know that that there is a consideration, like you say, of the child and and of of people's health, and of of people not coming to any harm, whether that harm is physical, or, you know, through up being upset or whatever. But generally I would say that after rigorous review, um, I don't think any research has ever been like completely stopped as such. I think it's more a case of you know. Really, this is something that you're going to have to rethink. Is there a way that you can? You can do that so I can see when you're saying that you know once you've explained it and justified it. That that would definitely be the case Workplace where you know the the environment is definitely, you know, this is important. Um. We need to find out about it. So therefore we'll support it. But you know you have to give us the the the details information. Yeah, Yeah, yeah, that that has, I think I mean as I said,
Ana Pellicer-Sanchez:H. Provided that i've, you know, given all the information. I've never had any any issues with, uh, you know approval. But it is an important concern, and that needs to be explained.
Nicole Brown:So some of the other questions that I've got is um A. Again, in terms of ethics. I've got one more question about ethics, and then i'm going to ask you a different one about the actual language learning. But um about the ethics. Um when you talking about, you know creating materials that are suitable and appropriate for the eye tracking because the text has to be a particular size, the lines have to be in a particular width. And yet, and all of that, what? How difficult is it to to create these materials without having a participant there, Um, or or do you? You know, sort of create them together with a a trial person, if you like. And how do you get that through ethical approval.
Ana Pellicer-Sanchez:Yeah, um, that's you. So that's a good point. I mean, usually. Of course, we we design these materials. It depends on the on the aim of the study, but of course trying to to mimic the kind of activities that they would do uh, or the kind of in this case the kind of text that they would be exposed to. Very, very often we start from an existing uh, you know, uh textbook uh uh um material, and then we modify it to make sure that if it's the uh, you know the the the requirements
Ana Pellicer-Sanchez: of the of the uh of the experiment uh always piloting it with either students, also with teachers. Uh, so the piloting of the materials is key, and in my experience it's never been um an issue with, uh, with ethical applications. Uh:sorry with ethics applications. It's always been Yeah, Um, You know the materials. I've always been
Ana Pellicer-Sanchez:very similar, if not the same, but similar to the kind of materials learners would only use in a normal context. So so it's never been an issue. Uh in my experience, I've never, you know, actually presented them very three key materials or anything but um, Yeah. So it's never been an issue. Actually, that's interesting. That's interesting, because my in my own research, I mean, I guess it's different, because i'm not using text. I'm using images and things. And sometimes when you use images, you know there's there's also of you know reasons as to why you're not supposed to be using images and what what you do. And you know, Yeah, it's probably in in that respect it's probably a different kind of It's more to do with with it. Not being a text, probably. I mean, we do have to be uh clear about whether there any sensitive topics that have been, you know, that are mentioned. Uh, of course you know for sure and say that you know none of the topics are sensitive, that or that culturally appropriate. That's another thing with with tests as well. Very often we have to make modifications depending on the country. We we run a particular study because we need to ensure that everything is culturally appropriate. So uh, so yeah, that's that's that's certainly something that goes in the ethics uh application form. I'm just laughing about that, because literally yesterday I recommended somebody some reading from the one thousand nine hundred and fiftys, and it's actually really inappropriate language. Wise, you know a piece of work that's been written in the S. It's kind of in a nowadays and modern day language. It's just really. Yeah, it would never get through now. Yeah. So the other question I've got is, I mean, this is more to do with the actual research that you. You were talking about. Um. You were obviously exploring the the the vocabulary learning in second second language learners.
Nicole Brown:Um, no, i'm wandering. I mean. You said early on that. There are about ten to fifteen percent of this a cost to our regression, and and and I totally get you know how that works, and all of that. But i'm kind of wondering uh the the regressions. How how can you know that the regressions happen? Because that child is second language learner, and not, perhaps, for example, a dyslexic, because a dyslexic child was also have more regression. So if somebody is perhaps not necessarily diagnosed with that dyslexia, um, you know they may have extra amount of regression. How how do you know? I mean? I'm not trying to throw open your case here, but i'm I'm. I'm. Generally interested in. How do you know that that is a um, a language learning issue and not a dyslexia issue?
Ana Pellicer-Sanchez:Yeah, that's a very good point. And there some research on that. How do we? What usually what we try to do is in in background questionnaires trying to get information about any any sort of uh, you know conditions that might affect your reading now. Um, sometimes, as you say, participants wouldn't know themselves, so there is no way we could know if a regression is actually uh, you know, because of, you know, a result, or even, you know, dyslexic readers might have a a longer reading times, or my fixate more on certain parts of the text. Uh, there is no way of knowing just by looking at the eye movement data. Really. Um, I mean the way to um, I mean. Yes, ideally. You would want to uh explore these population separately, and this is what they what this is, Why, they are studies on these lexic readers. Um, but um Usually in this type of studies where we have a large enough group, and we're talking about group patterns. If we have one or two participants who didn't know they actually were dyslexic, and they would still uh um. You know, I don't think that would affect the general patterns that much. Of course, if you have some of the participants being a undiagnosed uh uh, you know, dyslexic that would affect the patents. But we hope that's not. That's not really the case. If it's just one or two uh, you know. Just a few participants who don't know about it. Um, you know. Then uh, I don't think that would affect uh um a results that much. And this is why it's so important to have an enough sample size, so that there there is enough power in the analysis and the results. Um, I also tend to ask them um at the end of the experiment, how they felt about reading it, whether they found any sort of difficulties, because you never know, for not only for for and dyslexia, but for all the reasons I mean. Sometimes we will say, Oh, I really struggle to to find, you know, to read these texts, or I didn't really understand that that might for me there might be a reason for deleting, you know, data from that particular participants. So you know some sort of after um um reading, you know, Exit. Uh questionnaire, or questions are important, I think, as well as a way of screening that
Nicole Brown:that's really interesting. I mean, I i'm. I'm just particularly interested in that. Because um, I have got a child, a son who is dyslexic and bilingual. So we've got, and obviously in school he used to have French lessons as well. So we we. We have gone through all of those difficulties of of a multilingual brain at the same time Doesn't read very well, So it's It's It's something that's you know, genuinely of interest. And And yeah, we've we've had a lot of a lot of sort of issues and and tears around that when it was about having learning to read is quite,
Ana Pellicer-Sanchez: quite stressful for for for parents and children. Yeah, and it's Um:yeah, it's not yeah, not always understood properly, really.
Nicole Brown:And the thing is also what was interesting with him was Um, that at the time, you know, there were certain things that came through where it was very clear that you know his his understanding of language and of cultural co notations. Erez Agmoni was impeding some of his his basic reading skills. And that's that's really interesting here. So sometimes there was the the bilingualism that actually affected the the everyday learning one. Um! He He struggled to read the time in in English because he read it in German perfectly. Um! And and that's that's the kind of thing that we were really really struggling with at a time. You know that it's just when I don't know there was no logic behind it. You know this is the problem that was, and nobody could give you any guidance, because that was something that was new to everybody who was sitting there with in as a stakeholder in that relationship, and it's so individual as well, I mean monthlying will. Kids would
Ana Pellicer-Sanchez:not necessarily all do the same. So it's really really uh tricky, and we all know that in the end it obviously pays off, and the advantages overcome any sort of initial disadvantages, and they catch up, and all that. But of course, going through that process where you see your kids struggling. It's it's difficult. And this is why a lot of people move, move, move, move to. You know a a monolingual environment. And they say, Okay, yeah, let's forget about, you know, trying to to to maintain these different languages, or or you know which, of course we shouldn't do. We should. We should all aim for for for that. But it's more challenging. Absolutely.
Nicole Brown:Well, thank you very much, Anna. This was absolutely exciting. I have got one question here from the floor, so please do you want to come in, or do you want to raise the question in the chat box.
Ana Pellicer-Sanchez:Um, hi, Anna, do you hear me?
Sheng Zhang:Uh? First of all, i'd like to say Hi, Jenna, because uh I was a previous in a T. So in service student.
Ana Pellicer-Sanchez:I registered your uh so a course
Sheng Zhang:uh this, Webinar and I do have this question since you mentioned about a study about pre uh teach vocabulary. Now i'm a secondary English teacher in China, and um, you know, with very limited experience. But I feel like um, you know, regarding pre teach vocabulary. I feel like um. It is more beneficial to students who have a relatively higher level Proficiency? Um, I mean for higher level proficiency of students. It's better for them to learn students in context. So normally we won't choose uh to pre pre teach vocabulary. We're at lower level proficiency students. We tend to pre teach them so hopefully can lower their um cognitive. Um uh demands absolutely. I haven't read any papers about empirical studies about that to improve the the decision making. Um, So do you have any comments about this?
Ana Pellicer-Sanchez:The participants, and on their level of proficiency? So they are. Um, So you're right. This pre teaching of vocabulary. Reading, instruction of vocabulary is often use the minority of times to support learners who might have difficulties. Right reading that text with those particular words. So you we predict those difficulties and say, Okay, let's prepare these people Better let's teach uh this potentially. Challenge your vocabulary, and then we um, you know we um uh, we let them read the text. So that's one, and that's the kind of situation you were referring to, but it really depends on the context. So if we think of the Academy context, for example, even very, you know, high level uh users and and learners of of a language can be often be exposed. This is something that we do. Um um, and we see a lot in in, you know, in um a specialized vocabulary, for example, if you're learning, you know you're starting. University might be a very advanced user of of of of the of the language. But you might be completely new to this type of vocabulary. So actually, it's not uncommon in academic context to actually see, you know, typical glossary of terms. Uh, you know, at the beginning of the units before actually reading about the the topic. So in in in this context, actually, we might see examples of pre, you know, instruction of vocabulary or intentional learning of vocabulary before reading, even with more advanced learning. So I think it really depends on the on the topic, and we might use it for different purposes, depending on the sorry on the participants.
Nicole Brown:Thank you very much. Thank you very much, Anna. That was really really um very, very um helpful sort of explanation. I'm. I am just very quickly sharing um some of the information um on the Practice Research Network in case anybody would like to subscribe to the Newsletter um, or, you know, find out a little bit more. Um. Also, I would like to highlight that the next seminar is going to take place on the second of November. Um. Where, Dr. Elena, that I've always going to be talking about ethics, research practice to the Company Three is a crowd. So she's going to be exploring. And you know that kind of that that that we have when we're doing all of those things together, ethics, research, and practice. Um, so can I just say, Anna, Thank you very, very much for today. Thank you. Super exciting. It was absolutely interesting. I can I can not. You know. I'm. I'm generally pleased that you've been here, and that we've had an opportunity to talk about your work. And and yeah, I look forward to to hearing more in in the future. So much. Thanks a lot, Nicole. Thanks a lot for the invitation. So yeah, yeah, yeah, And thanks thanks to everyone. Thanks a lot. My pleasure. Thank you. And see you soon. Yes,