欢迎来到留学生英语论文网

客服信息

我们支持 澳洲论文代写 Assignment代写、加拿大论文代写 Assignment代写、新西兰论文代写 Assignment代写、美国论文代写 Assignment代写、英国论文代写 Assignment代写、及其他国家的英语文书润色修改代写方案.论文写作指导服务

唯一联系方式Q微:7878393

当前位置:首页 > 论文范文 > Psychology

Biological and Cognitive Psychology Visual Passwords Experiment

发布时间:2017-04-06
该论文是我们的学员投稿,并非我们专家级的写作水平!如果你有论文作业写作指导需求请联系我们的客服人员

The aim of the current study was to examine how well (accurately) people recall different types of images, and how quickly they recall the images. These are to see which images are best remembered and recalled, and would make easier-to-remember visual passwords. A total of 39 undergraduate participants from mixed genders were recruited from the University of Bedfordshire. The stimuli were presented through a server. 45 images of each type arranged in arrays of 3x3. Participants had to experience three different phases: choice phase, training phase, and recognition phase. The recognition phase was repeated 4 weeks after the initial learning to examine long term recall of the images. The finding associated that people remembered images of faces more as compared to images of abstracts and objects. Results also indicated that they recalled better for image of faces after 4 week delayed. Hence, hypotheses supported.

Challenge a user with a seemingly unique alphanumeric password is the most general paradigm. However, there are many hidden problems on alphanumeric passwords. A password must be changed regularly to be very secure for compounding the problem for a user to recall. On the other hand, something where people have to distinguish an image from a series of images is visual passwords. What is ambiguous is the type of image to be used in such password system.

Besides that, object recognition involves between- category discriminations. Grill- Spector and Kanwisher (2005; as cited in Robertson, 2009) have shown that we recognize what an object is (its category) as soon as we know it is there. Also, accessing the semantic system (information related to the object is stored including its name) is needed when recognizing an object. Therefore, there are multiple sources of information in encoding and recognizing a formerly seen object. Abstract images do not look like any real objects. As a result, the least well recognized will be abstract images.

Cartens, McCauley-Bell, Malone, & DeMara (2004) conducted a research on evaluation of the human impact of password authentification practices on information security. Researchers created a model to prophesy the vulnerability of a certain set of conditions would have on a probability of error in an information system via distributing a survey and conducting an experiment. A password information security survey was distributed over 250 respondents and a limited case study consisted of 30 participants was performed. The experiment contained three stages that examined different password difficulty levels. Participants are required to choose their own passwords that satisfied stringent password guidelines, whereas the second stage and third stage required participants to form their passwords by using chunking meaningful data together.

The survey results indicated that those individuals with eight to eleven work or school passwords are at greatest risk for not remembering their passwords at least once a month. The results of the experiment indicated that passwords had a significantly higher level of remembrance at second and third stage. Hence, these findings indicated that the use of passwords comprised of data meaningful for the user can significantly reduce the human error associated with password authentification and that meet the information technology community requirement for strength of password.

Brostoff and Sasse (2000) conducted study on passfaces. They were interested on investigating whether a new mechanism for user authentication- passfaces are more usable than passwords. According to Brostoff & Sasse (2000), the proliferation of technology requiring user authentification has increased the number of passwords which users have to remember, creating a significant usability problem. Therefore, a usability comparison between passfaces and passwords, with a number of 34 students participated in a 3-month field trial were reported. A repeated measures design was used, with each participant using both passwords and passfaces. Then, logins and login problem rate were analyzed, followed by reminders, time taken before first use of the system, and number of logins. Brostoff & Sasse (2000) reported "fewer login errors were made with Passfacesm even when periods between logins were long". Passfaces took a long time to execute on the computer facilities chosen by participants to log in. The importance of evaluating the usability of security mechanisms in field trials were also emphasized by the results.

Nagata (1986) was interested in examining a study of picture recognition of concrete common objects and abstract random shapes under two presentation conditions: the duration of the unfilled intervals (off-time duration) that followed the presentations of the pictures were varied either randomly or in blocks. A total of 36 students were recruited. Concrete and abstract pictures were used for this experiment. Nagata (1986) reported the major findings were: (a) recognition of the concrete pictures was better than recognition of random shapes, (b) the off-time duration did not influence recognition of either type of picture under the random presentation condition, and (c) under the block presentation condition, recognition of the concrete pictures increased as a function of off-time duration, whereas the effect of off-time duration was not found for the abstract shapes. Hence, recognition was better for concrete pictures than for abstract random shapes.

Thus, the present study was to ascertain how accurately people recall different types of images, and how quickly they recall the images. These are to see which images are best remembered and recalled, and would make easier-to-remember visual passwords. Hence, the first hypothesis which outlining the present study is that there would be a significant main effect of phase on mean reaction times. The second hypothesis is that the abstract images are significantly harder to remember after 4 weeks delayed compared to the images of faces and object.

Method

Design

A 3x2 repeated measures of design was employed for current study: image types (faces, abstract, and object) x phases (immediately and delayed). The number of images correctly recalled by the participants is the dependent variable (DV), whereas the independent variable (IV) are the category of images and the phases (immediately and delayed). In order to eliminate confounding variables, the sequence of images appear were counterbalanced with 6 different orders for each participant which are 1) abstracts, objects, faces (AOF); 2) AFO; 3) OFA; 4) OAF; 5) FAO; and 6) FOA.

Participants

A total of 39 undergraduate students recruited from the University of Bedfordshire. Participants were currently enrolled in second year of Psychology and consisted of 16 males and 32 females with the mean age of 23.

Materials

There were 15 images of each type arranged in arrays of 3x3. Hence, there were 5 sets of arrays of 3 image types (faces, objects, and abstract images). (Please refer to Appendix A for examples.) All images were the same overall size (3cm2). The faces and objects had a white background whilst the abstract images filled the 3cm square. The experiment was programmed and run using Superlab© version 2.04 on and IBM/PC compatible computer running Windows XP Professional© and displayed on a 17" flat screen monitor.

Procedures

These formed the stimulus type which is the independent variable. Other than that, the time participants had to remember the images (immediately versus delayed) was varied, and this is the second independent variable. Apparently, the number of items correctly recognized would be measuring. However, we are also interested whether one condition is harder than another, so reaction time will be taken as well. So, there were two dependent variables- correct responses and reaction times, therefore two types of analysis will be carried out. The statistical test used is 3x2 repeated measures two-way ANOVA. There were three phases which participants had to experience. The very first phase is the "choice phase". An array of images was presented; participants were given the chance to choose one of them. Four more times were repeated to allow the participants to select five images that represent his or her "password". The next phase is called "training phase". By recognizing the images he or she had chosen, participants were given the opportunity to train themselves amongst a set of "distractor" images. After that, the image they have chosen was highlighted in yellow, which they were required to click on it. Both the target and the distractors, were randomly re-arranged in this phase. The following phase is the "recognition phase". Participants were then presented with the same arrays with the individual images again randomly re-arranged after both images types had been shown. At this time, participants had to recognize their image which means the "target" amongst the distractors. The recognition phase was repeated four weeks after the initial learning to examine long term recall of the images. Data were recorded after the experiment and analyzed using two-way ANOVA by SPSS program.

Results

Results were derived from the participants. The first session of the results is the number of items correctly recalled for both immediate phase and delayed phase.

Table 1

The descriptive data on the number of correctly recalled items on immediate phase and delayed phase.

Mean

Std. Deviation

N

FACE- IMMEDIATE

4.9231

.35427

39

ABSTRACT- IMMEDIATE

4.6667

.57735

39

OBJECT- IMMEDIATE

4.9487

.32026

39

FACE- DELAYED

4.4872

.68333

39

ABSTRACT- DELAYED

3.6154

1.09100

39

OBJECT- DELAYED

4.5128

.75644

39

Repeated measures of an analysis of Variance (ANOVA) were performed to test these means for main effects of immediate phase and delayed phase with image types (faces, abstract, object). The mean of number of items recalled correctly for immediate phase was 4.83, while the mean of number of items recalled correctly for delayed phase was 4.2. Hence, there was a significant main effect and difference between immediate phase and delayed phase, F (1, 38) = 45.382, p< 0.05. On the other hand, the condition means for faces, abstract, and objects were 4.9, 4.2, and 4.7 respectively. Therefore, there was a significant main effect of image type on number of items correctly recalled, F (2, 76) = 20.583, p< 0.05.

Table 2

The mean data of pairs of image type.

(I) imgtype

(J) imgtype

Mean Difference (I-J)

Sig.a

1

2

.564*

.000

3

-.026

1.000

2

1

-.564*

.000

3

-.590*

.000

3

1

.026

1.000

2

.590*

.000

*. The mean difference is significant at the .05 level.

a. Adjustment for multiple comparisons: Bonferroni.

There was a mean difference between the images of faces (1) and abstract (2) at 0.564, p< 0.05, which is significant. There was a mean difference between the images of abstract (2) and object (3) at -0.590, p<0.05, which is significant. However, the mean difference between the images of object (3) and faces (1) was 0.026, p> 0.05, which has no significant difference. The graph below illustrates that the number of image of faces and objects were recalled correctly more than the images of abstract for both phases.

Graph 1. Ilustrative statistics for ANOVA for number of items (image types) recalled correctly in immediate phase (1) and delayed phase (2).

In addition, the interaction between phase and image type on the number of items recalled was also examined. Results indicated that there was a significant interaction between phase and image type, F (2, 76), = 7.664, p< 0.05.

Table 3

The data of the main effect of phase for each type of image.

IMG TYPE

(I) PHASE

(J) PHASE

Mean Difference (I-J)

Sig.a

1

1

2

.475*

.000

2

1

-.475*

.000

2

1

2

1.000*

.000

2

1

-1.000*

.000

3

1

2

.450*

.001

2

1

-.450*

.001

Meanwhile, there were also simple main effects of image type by phase of recall. The table above illustrates that the mean number of faces (1) recalled immediately is 0.475 higher than the mean number recalled after a 4 week delay, and that this difference is significant (p<.001). The mean number of abstract (2) recalled immediately is 1 higher than the mean number recalled after a 4 week delay, and that this difference is also significant (p<.001). Lastly, the mean number of object (3) recalled immediately is 0.450 higher than the mean number recalled after a 4 week delay, and that this difference is significant (p<.001) too. Hence, the second hypothesis is supported by the first session of the results, which is that the abstract images are significantly harder to remember after 4 weeks delayed compared to the images of faces and object. To explain this, the reason might be the detail of the images of abstract is not clear, it seem all the same as compared to faces and objects.

The second session of the results is the reaction time data for immediate phase and delayed phase.

Table 4

The descriptive data of reaction time taken for immediate phase and delayed phase.

Mean

Std. Deviation

N

FACE- IMMEDIATE

12703.5897

7246.39548

39

ABSTRACT- IMMEDIATE

20382.8974

9713.97374

39

OBJECT- IMMEDIATE

12807.8462

12889.29054

39

FACE- DELAYED

25543.7949

10669.28436

39

ABSTRACT- DELAYED

31593.3590

13785.75312

39

OBJECT- DELAYED

24236.3333

11467.45214

39

Repeated measures of an analysis of Variance (ANOVA) were performed to test these means for main effects of immediate phase and delayed phase with image types (faces, abstract, object). The mean of number of reaction time for immediate phase was 15298.1, while the mean of reaction time for delayed phase was 27124.5. Hence, there was a significant difference and main effect of phase on mean reaction times, F (1, 38) = 43.925, p< 0.05. On the other hand, the condition means for faces, abstract, and objects were 19123.7, 25988.2, and 18522.1 respectively. Hence, there was no significant main effect of image type on reaction times, F (1, 38) = 0.085, p> 0.05. The graph below illustrates the means of reaction times taken for both phases on each image type.

Graph 2. Ilustrative statistics for ANOVA for reaction time taken for each image type in immediate phase (1) and delayed phase (2).

The interaction between phase and image type on reaction time was also examined. Results indicated that there was no significant interaction between phase and image type, F (1, 38), = 0, p> 0.05. Simple main effects for this session will not be presented as there was no significant interaction between phase and image type on reaction time. Hence, the first hypothsis is supported by this session which is that there would be a high significant main effect of phase on mean reaction times.

Discussion

Exploring at the pattern of the results from the current study, there was a main effect and difference between immediate phase and delayed phase for the number of items correctly recalled. People tend to forget certain thing after a period of time. Moreover, results also indicated that image type has an effect of on number of items correctly recalled as image of faces is highly recalled after 4 week-delayed whereas abstract images were least remembered. Research done by Brostoff and Sasse (2000) on investigating whether passfaces are more usable than password was related with the section of number of items correctly recalled in this current study as they found that fewer login errors were made with Passfacesm even when periods between logins were long. In addition, table 3 showed that the mean difference of abstract (2) recalled immediately after a 4 week delayed is the highest amongst three image types, this supported the second hypothesis which is that the abstract images are significantly harder to remember.

Participants remembered image of faces more might be the reason that the images used are multi-cultural of faces, consists of Western faces and Eastern faces. The features of faces are easier to be recognized and recalled than abstract images because faces could be in a variety of colors, skin tones and shapes. Object images on the other hand also highly recalled by participants compared to abstract images, it might because of the object in the images are common things in our everyday life such as cups, gloves, battery, and etc. There might also be a reason on the differences on individual's memory.

However, there is a confounding variable underlying in this study, which is that the genders might have an effect on the results because it was not equalized. Having said that, the present study is successful in some cases because hypotheses are supported and results are what to be expected. For future research, age group and gender differences could be taken into account on similar study because memory might be different between both genders and different age group.

上一篇:The prenatal influence of rubella on development 下一篇:Internal And External Sources Of Individual Pressure Psychology Essay