<?xml version="1.0" standalone="yes"?>
<Paper uid="W98-1229">
  <Title>Towards Language Acquisition by an Attention-Sharing Robot</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
2 Attention-Sharing and
</SectionTitle>
    <Paragraph position="0"/>
    <Section position="1" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
Symbol Acquisition
</SectionTitle>
      <Paragraph position="0"> Observation of others' verbal behavior provides infants with learning data for symbol acquisition. Let us consider that an agent, looking at a cat, says &amp;quot;cat&amp;quot;, as illustrated in Figure 2. In order to mimic this verbal behavior, the self has to observe the  (1) capture gaze Figure 1.</Paragraph>
      <Paragraph position="1"> (2) identify the target Attention-sharing based on gaze.</Paragraph>
      <Paragraph position="3"> agent's sensory-input I (stimulus from the cat) and motor-output O (verbal response) and to make the association between them.</Paragraph>
      <Paragraph position="4"> Attention-sharing enables us to observe someone else's input and output, as also shown in Figure 2. Attention-sharing guarantees that I' (the self's input) resembles I, since both are paying attention to the same target. At the gaze-capturing stage (Figure 1, left), the self can observe the agent's output O and map it onto the self's motor image Oq (We assume an innate mapping mechanism like imitation of facial gestures by neonates.) Although thus observed relationships between input space and output space may vary in many ways (size, color, tone, volume, etc.), one can construct an efficient mediator between these spaces. As illustrated in Figure 3, the complex relationships (left) Kozima and 1to 245 Language Acquisition by an Attention-Sharing Robot  can be decomposed into several (almost separated) components by introducing a hidden mediator space M (right), on which &amp;quot;symbols&amp;quot; can emerge.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>