BCS152-Tutorial/SPRDesign.html at master · hlplab/BCS152-Tutorial · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
<!DOCTYPE html>
<html>

  <head>
    <meta charset="utf-8">
    <meta name="viewport" content="width=device-width, initial-scale=1">
    <title>BCS 152 Tutorial - Design</title>
    <link rel="stylesheet" href="//maxcdn.bootstrapcdn.com/bootstrap/3.3.4/css/bootstrap.min.css">
    <link rel="stylesheet" type="text/css" href="tutorial.css">
  </head>

  <body class="bckgrnd">
    <nav id="nav" class="navbar navbar-default navbar-fixed-top"></nav>

        <div class="jumbotron spr top">
          <h1>Self Paced Reading Example: Design</h1>
            <p>
              [<a href="#work" class="bldspr">Working hypothesis</a>]
              [<a href="#2x2" class="bldspr">A 2x2 Example Design</a>]
              [<a href="#worb" class="bldspr">Within- or Between</a>]
            </p>
            <p>
              In this section, we expand on the example experiment we've introduced
              above, which manipulated word frequency. We extend this example of a by-2
              design to 2x2 design (read: "two-by-two design").
            </p>

          <p id="work">
            But first things first. We'll follow the guidelines described so far
            and begin by outlining a research hypothesis that we want to test.
          </p>
      </div>

      <div class="jumbotron spr">
        <h2>Working hypothesis</h2>
          <p>
            For this example experiment, we start with the hypothesis that language
            processing is
            <a href="Glossary.html#E" target="_blank" class="red" data-toggle="tooltip" title="">expectation-based</a>.
            That is, we hypothesize that
            <a href="Glossary.html#C" target="_blank" class="red" data-toggle="tooltip" title="">comprehenders</a>
            automatically draw on implicit statistical knowledge about the distribution
            of linguistic elements while processing language. Although we will not
            further focus on it this year, this hypothesis is in turn derived from a
            more general theory&mdash;namely, that the brain makes rational
            use of available information to robustly and efficiently infer linguistic
            structure (and thereby sentence interpretations) from noisy perceptual
            input.
          </p>

          <p>
            To operationalize this hypothesis, we will narrow it down further.
            Specifically, we plan to test whether expectations about upcoming
            <em>words</em> affect how fast these words are <em>read</em>. That is,
            we are testing the expectation-based hypothesis with regard to expectations
            about words while using reading (rather than, for instance, listening or
            viewing sign language) as the mode in which language is presented.
          </p>

          <p id="2x2">
            If the expecation-based hypothesis is correct, words that are more expected
            should be read more quickly and words that are less expected will be read
            more slowly. It is also possible that some of the effects of processing
            difficulty due to a word's expectedness will
            <a href="Glossary.html#S" target="_blank" class="red" data-toggle="tooltip" title="Spill-over means that the processing difficulty associated with a word leads to slower reading times on the immediately subsequent words.">spill over</a>,
            thus affecting the reading times of subsequent words. With this in
            mind, we are now ready to describe the design we will use to test
            this prediction.
          </p>
        </div>

        <div class="jumbotron spr">
          <h2>A 2x2 Example Design</h2>
            <p>
              We will test our working hypothesis in a 2x2 design, by crossing word
              frequency (high vs. low frequency) with contextual predictability (high
              vs. low predictability). This design is illustrated by Figure 1, which
              shows an example of a critical item in its four (2x2) conditions.
            </p>
            <figure>
              <img title="Example 2x2 Design" alt="Example 2x2 Design" src="Example.png">
              <figcaption style="text-align:center;">Figure 1: A 2x2 design, crossing contextual predictability (high vs. low) with word frequency (high vs. low).</figcaption>
            </figure>
            <br>

            <p>
              This design allows us to test the expectation-based hypothesis in two ways.
              First, if a word is in a high predictability context (e.g., "apple" or
              "kiwi" after the verb "ate") then it should be read more quickly.
              Second, if a word is in a low predictability context (e.g., after the
              verb "saw"), then more frequent words (e.g., "apple") should be read
              more quickly than low frequency words (e.g., "kiwi").
            </p>

          <p id="worb">
            That is, we might predict contextual predictability and word frequency to
            <a href="Glossary.html#I" target="_blank" class="red" data-toggle="tooltip" title="">interact</a>,
            with frequency effects being larger (or even only present) in the low
            predictability condition. Alternatively,
            we might find that contextual predictability and word frequency have
            independent additive effects on reading times.
          </p>
        </div>

        <div class="jumbotron spr">
          <h2>Within- or Between-?</h2>
            <p>
              The design in Figure 1 is an example of a within-item design: across
              participants, each critical item occurs in all conditions (this does
              not mean that every single participant sees each item in all of its
              conditions; more on that below).
            </p>

            <p>
              With regard to participants, we could conduct the experiment as within-
              or between-participant design. For now, we will
              take the simplest option and make the design within-participant.
            </p>

          <p>
            Of course, we don't want to run an experiment with just one item.
            As we've discussed above, that wouldn't let us generalize from whatever
            we find in our experiments to how people process language more generally.
            Instead, our experiment will have many items, all of them following
            the same structure of the example item shown in Figure 1. In
            later sections, we will cover more details about items and then
            go through how we can construct lists of items for our experiment. Before
            that we will go over the self-paced reading procedure we will use to
            collect our data.
          </p>
      </div>

      <div class="jumbotron">
        <div class="container">
          <div class="row">
            <table>
              <tr>
                <td class="col-md-12"><div><a class="btn btn-primary" href="DesignTypes.html">&larr;Types of Design</a></div></td>
                <td class="col-md-12"><div><a class="btn btn-warning" href="SPRProcedure.html">SPR: Procedure &rarr;</a></div></td>
              </tr>
            </table>
          </div>
        </div>
      </div>

        <!--<div class="jumbotron">
          <button type="button" class="btn btn-block btn-lg btn-danger" data-toggle="collapse" data-target="#demo">Test Your Understanding</button>
            <div id="demo" class="collapse">
              <p class="question">1. What is an item?</p>
              <ul class="answers">
                <li><input type="radio" name="q1" value="a" id="q1a"><label for="q1a">Another word for the general term "stimuli"</label></li>
                <li><input type="radio" name="q1" value="b" id="q1b"><label for="q1b">One of the requirements for a specific condition</label></li>
                <li><input type="radio" name="q1" value="c" id="q1c"><label for="q1c">A set of examples that fulfills the requirements of each condition</label></li>
              </ul>

              <p class="question">2. What is a list?</p>
              <ul class="answers">
                <li><input type="radio" name="q2" value="a" id="q2a"><label for="q2a">A sequence of examples representing each of the conditions</label></li>
                <li><input type="radio" name="q2" value="b" id="q2b"><label for="q2b">A set of conditions for particular groups of subjects </label></li>
                <li><input type="radio" name="q2" value="c" id="q2c"><label for="q2c">The group of factors and variables that we are testing in the experiment</label></li>
              </ul>
              <p class="question">3. True or False: In our case, each condition must have 6 to 8 items.</p>
              <ul class="answers">
                <li><input type="radio" name="q3" value="a" id="q3a"><label for="q3a">True</label></li>
                <li><input type="radio" name="q3" value="b" id="q3b"><label for="q3b">False; each item must have 6 to 8 conditions</label></li>
              </ul>
              <div id="results">
                Check
              </div>
              <div id="alerts"></div>
            </div>
        </div> -->

    <script type="text/javascript" src="//cdnjs.cloudflare.com/ajax/libs/jquery/2.1.4/jquery.min.js"></script>
    <script type="text/javascript" src="//cdnjs.cloudflare.com/ajax/libs/underscore.js/1.8.3/underscore-min.js"></script>
    <script type="text/javascript" src="//maxcdn.bootstrapcdn.com/bootstrap/3.3.4/js/bootstrap.min.js"></script>
    <script type="text/javascript" src="testnav.js"></script>
    <script type="text/javascript" src="sprdesign.js"></script>
    <script type="text/javascript">
      $(document).ready(function() {
        $('li#design').addClass('active');
        $('li#sprdes').addClass('active');
      });
    </script>
  </body>
</html>