Spring 2019
Hello Parents! We are grateful to the parents and children who have participated in our studies, and those who have spread the word about our research program! Your participation and support have added to our understanding of child development, and have made important contributions to the larger scientific community; our research has been published in academic research journals and presented at professional conferences. This newsletter presents an update regarding some of the studies that we have previously conducted or are actively pursuing.
For those parents who may not have been in our studies (yet!), we are located on Binghamton University's campus and our lab, the Infant and Child Studies Project, has been investigating visual and cognitive development since 1997. Our studies use eye-tracking technology, touchscreens, videos, closed-circuit TV (CCTV), and puzzles to engage children's attention and ensure they have fun while contributing to science! We have a newly renovated lab on campus, with a playroom filled with lots of toys and a great group of students who are excellent with kids. Our studies typically last between 30-60 minutes and appointments are scheduled at your convenience (weekday and weekend appointments are available).
We hope that you will enjoy this newsletter. If you are interested in participating in one of these studies, please do not hesitate to reach out to us at baby@binghamton.edu. As always, none of this would be possible without your continuing support. Thank you!
Learning from Screen Media in Children
Technology and screen media is an increasing source of interest and concern in our society. It is a little difficult to believe that it has been less than 9 years since the first iPad was introduced. The context in which young children engage in learning is changing with the technological revolution. More than 80% of apps are directed at children, and parents and educators are told that screen media can be beneficial for learning. Adults may be fully able to learn and transfer information to-and-from the virtual environment, but children are less able to acquire and apply (that is, transfer) knowledge between dimensions. We have found that children learn less when they are tasked with taking information from a screen and applying it to the real world, and vice versa, compared to traditional face-to-face learning. If you'd like to know more about this topic, check out some of our recent publications. In current studies, we are investigating this learning deficit specifically in toddlers, and we are particularly interested in how the social environment impacts their ability to learn in a virtual environment.
In one study, 2.5 to 3-year-olds watch a series of short videos presented on a TV monitor. The videos depict two events; a three-piece puzzle being assembled and a distractor object. In the center of each video, an actress provides varying degrees of social support: (1) speaking directly to the child, (2) speaking, but not to the child and/or (3) shifting her eye-gaze towards the puzzle. These social cues have been shown to increase visual attention to the task and facilitate learning. The task for this study is learn how to assemble the puzzle. We use an eye-tracker to determine precisely where the child is looking on the screen while they are watching the video. Then, the child is given a 2D touchscreen (iPad) version of the puzzle and we observe how they interact with the puzzle. We are finding that children spend the most time looking at the puzzle compared to the distractor object and the actress, which is a pattern most evident in 3-year-olds. These older children are better at completing the 2D puzzle compared to 2.5-year-olds. The younger age group appears most successful at assembling the puzzle when the actress in the video speaks, directly or indirectly, compared to children who see less socially engaging (e.g., eye gaze only) videos. This suggests that speech is highly attention-grabbing and may increase overall attention during the task, and subsequently, aid learning.
We will soon begin running a separate, but related, study with mother-child pairs. In this task, moms will be familiarized with a 3D magnetic puzzle or a 2D virtual puzzle and asked to teach their child how to assemble the puzzle. This learning environment is complex, with many social cues, so lightweight head-mounted cameras will be used to capture looking behaviors, gestures, and/or verbal cues between mother and child during this short interaction. We hope to see what similarities and differences these two learning environments produce. This should help us to understand why children learn less from screen media and what aspects of the social setting might help them learn better! We are still looking for 1.5- to 3.5-year-olds for these two studies! Please contact us if you and your child would like to play our puzzle games!
We are still looking for 1.5- to 3.5-year-olds for these two studies! Please contact us if you and your child would like to play our puzzle games!
Learning from Screen Media in Adults
We know that infants and young children have difficulty transferring learning between screen media (e.g., television and touchscreens) and the real world, but does this transfer deficit persist in adults? Most of us would suggest that it does not; however, this is not necessarily the case.
We tested this question by having adults (college students) observe the assembly of a 7-piece multi-color, 10-piece multi-color, or a 7-piece single color puzzle, with either a 3D physical puzzle or on a 2D touchscreen (iPad). They were then asked to assemble the puzzle that they had seen demonstrated (e.g. if a 7-piece single color puzzle was demonstrated, they were asked to complete a 7-piece single color puzzle). Participants may have had to assemble a puzzle in the same dimension they saw demonstrated (e.g. 3D demonstration, 3D puzzle to assemble) or in a different dimension (e.g. 2D demonstration, 3D puzzle to assemble).
No evidence of a transfer deficit was found with easier puzzles; participants were able to assemble the puzzle equally well whether their demonstration matched or differed from the dimension in which they assembled the puzzle. Participants with the most difficult puzzle (7-piece single color) performed poorly if the demonstration and test dimensions differed. Thus, when the task is difficult, there is some evidence of a slight deficit in transferring learning to and from screen media in adults.
In our tasks, the difference between the 3D and the 2D puzzles was minimal; they were designed to be as similar as possible! However, in many real-world learning environments, there is a larger difference between what one sees on a touchscreen and what one sees in the 3D world. We will soon extend this test to more realistic situations, given that this study has implications for our understanding of adults' ability to learn from screen media in multiple contexts, such as online education.
Illusory Contours
Adults have a remarkable ability to "fill in the blanks" when something is not entirely visible, a process called visual interpolation. However, we are not born with the recognition skills that we have as adults; these skills develop through childhood. We can test how children and adults interpret objects with missing visual information by using illusory contours, such as the one shown above. Illusory contours are simple shapes that have incomplete boundaries. The Kanizsa "square" (which we use in our study) is made up of four 'pacmen' shapes as shown above. Even though there are no physical lines connecting one 'pacman' to the other, we still perceive a square!
This "square" is placed within an array of other randomly rotated pacmen and the task is to find the square as quickly as possible. It is harder to perceive illusory shapes depicted using smaller pacmen because there is less visual information (and thus more empty space) between the pacmen. You probably find it more difficult to find the square in the left picture than in the right picture.
We use eye-tracking to monitor where our observers are looking as they view these shape displays. We are interested in whether children in this study will be immediately drawn to the illusory contour, or if they have to search the screen to find it. As expected, we have found that adults tend to be faster and more accurate at finding smaller illusory contours than children.
We are currently looking for children ages 4-7 to participate in this study. Please reach out to the lab if you would like your child to have fun and "Find the square"!
Global/Local
When adults are shown images, they first look at the overall (global) scene, and then focus in on details (they "see the forest before the trees"). However, many studies have shown that children under the age of six notice the local details in an image at the expense of the global scene; this indicates that they are locally biased. Due to differences between previous studies, it is still unclear at what age children begin to notice global aspects over local features, which is why we are investigating this effect in 3- to 6-year-olds. We have determined that children's performance is affected by the type of instructions that are used, so we are investigating visual performance in children before and after both verbal and visual instructions.
During the study, children are shown a series of two objects side-by-side on a computer monitor. The objects either match completely (above, left), or they differ on either the global (center) or the local level (right). Children are asked to use a gamepad control to indicate whether the shapes are the same or different.
Halfway through the experiment children are given a verbal and visual instructional sequence detailing what is meant by a local and a global difference to see whether drawing attention to both the global and local levels will affect responding. Before the specific instructions, 4-year-olds appear to notice the local difference, but not the global. After the instructional sequence, children are more likely to notice the global difference and less likely to notice the local. This suggests that visual bias towards whole or parts-based aspects of an image is more of a preference, and can be influenced and changed by simply drawing attention to certain aspects of an image.
We are looking for children ages 3-6 to participate. Please contact us if you would like to learn about what your child perceives as the same and different!
Mid-Vision Object Recognition
What remains after you can no longer see an object? What is the content of your memory for objects that allows you to recognize it again later on? The specifics of what is remembered is the focus of substantial research; do we remember every little detail, or do we only store larger units after the object is no longer visible? Research on adults has clearly shown that both feature-based (the straight vertical and curved horizontal lines and vertices that make up the body and lid of a grand piano, for example) and part-based (the body and leg of a grand piano, for example) descriptions are stored. Children, however, may represent objects differently. We are investigating whether children maintain the same set of multiple representations, or whether their memory tends to emphasize one type of information over the other.
So far, our studies have shown that young children have difficulty recognizing objects that are missing either parts or features. Unlike adults, missing features were more detrimental to children. These results suggest that children represent objects in memory differently than adults and rely more heavily on features, while adults are able to use both parts and features to extract object information.
Differential Outcomes Effect
Humans (and animals!) learn faster when they are given unique rewards for performing different tasks. Past studies have shown that this effect (called the Differential Outcomes Effect) occurs in both adults and children, with the effect emerging as early as 4 years of age and possibly earlier.
We are currently studying this effect in children between 4 and 6 years of age. We record children's eye movements in response to simple shapes shown on a television. The direction they look shows whether they have learned the two "rules". A look in the correct direction results in the appearance of a short movie scene. Some children who participate get two unique rewards (a movie for looking one direction and a different movie for the other) for looking in the correct directions (showing they have learned the "rules") and others get only one reward (one movie or the other) for looking in the correct directions. Children who see two unique rewards are predicted to learn the "rules" faster than children who receive only one reward each time they look to the correct location.
Much of the previous research exploring this effect in children has not been suitable for children under 4 years due to various aspects of the tasks, such as requiring children to understand complex verbal instructions, undergo extensive training, and/or make overt actions to respond. Eye-tracking is a useful tool for measuring visual working memory in young children and may effectively overcome such limitations. Therefore, this task is designed with as little spoken instruction as possible in order to determine whether this effect emerges earlier in development than previously reported! This will help us determine how children's learning changes or does not change when they have to figure out the task on their own.
So far, we have seen that the Differential Outcomes Effect occurs in young children, though to a lesser extent than it does in adults. In our eye-tracking studies of this effect, children tend to look only at the location at which they initially saw the movie. It may be that they do not understand that their eyes are actually doing something, so we are moving this experiment to a touchscreen to make it more clear that their actions make things happen.
We are looking for children ages 4-6 to participate. If your child enjoys listening to songs from popular children's movies, contact us for more information!
Line Orientation and Digital Media
How well do people detect the orientation of lines in a difficult setting? This apparently boring question is actually quite important because line detection is known to be crucial for object detection. Further, it is well understood that our visual system is not equally sensitive to all orientations; we can detect horizontal lines better than vertical, and vertical better than oblique (diagonal) lines. This is called the oblique effect. We have conducted a study using eye-tracking to test the impact of certain types of digital media exposure on this basic, low-level part of perception.
We did this by first testing the orientation detection skills of a group of college students; they proved to be typical observers, showing the oblique effect, as we expected (horizontal best, oblique worst). We then had them play a highly popular online game for four hours continuously and then tested them again.
Compared to a group that did not play the game (or have any digital media exposure) during the 4-hour period, the game players showed a small, but important shift in their sensitivity to line orientation; their ability to detect horizontal and vertical lines was enhanced, while their ability to detect oblique lines did not change (or may have declined just a little bit).
This is important because it suggests that playing the game just once can have an effect. We will soon recruit a number of children to do this as well, to see if the game affects their line orientation sensitivity in a similar manner. This study has important implications for understanding how exposure to screen media may alter basic aspects of our visual perception.
