"... In PAGE 6: ... 3.1 The Effect of Contexts on Human Behavior Table1 presents the average offer benefit to participants in both task and table con- dition for each role designation. Table proposers offered significantly more benefit to Table 1: Average Benefit of Offer... In PAGE 7: ... For the same round, this exchange may be different for task and table conditions. As shown in Table1 , there was no difference in the ratio of exchanges accepted by responders (77%) between conditions. However, this result does not mean that responders were not affected by context; as also shown in Table 1, they were responding to exchanges that were more helpful to them in the task condition.... In PAGE 7: ... As shown in Table 1, there was no difference in the ratio of exchanges accepted by responders (77%) between conditions. However, this result does not mean that responders were not affected by context; as also shown in Table1 , they were responding to exchanges that were more helpful to them in the task condition. We expected this pattern to hold for accepted offers as well; thus, we expected that the offers that were accepted by responders were more helpful to them in the task condition than in the table condition.... ..."

"... In PAGE 6: ... 3.1 The Effect of Contexts on Human Behavior Table1 presents the average offer benefit to participants in both task and table con- dition for each role designation. Table proposers offered significantly more benefit to Table 1: Average Benefit of Offer... In PAGE 7: ... For the same round, this exchange may be different for task and table conditions. As shown in Table1 , there was no difference in the ratio of exchanges accepted by responders (77%) between conditions. However, this result does not mean that responders were not affected by context; as also shown in Table 1, they were responding to exchanges that were more helpful to them in the task condition.... In PAGE 7: ... As shown in Table 1, there was no difference in the ratio of exchanges accepted by responders (77%) between conditions. However, this result does not mean that responders were not affected by context; as also shown in Table1 , they were responding to exchanges that were more helpful to them in the task condition. We expected this pattern to hold for accepted offers as well; thus, we expected that the offers that were accepted by responders were more helpful to them in the task condition than in the table condition.... ..."

"... In PAGE 6: ... 3.1 The Effect of Contexts on Human Behavior Table1 presents the average offer benefit to participants in both task and table con- dition for each role designation. Table proposers offered significantly more benefit to Table 1: Average Benefit of Offer... In PAGE 7: ... For the same round, this exchange may be different for task and table conditions. As shown in Table1 , there was no difference in the ratio of exchanges accepted by responders (77%) between conditions. However, this result does not mean that responders were not affected by context; as also shown in Table 1, they were responding to exchanges that were more helpful to them in the task condition.... In PAGE 7: ... As shown in Table 1, there was no difference in the ratio of exchanges accepted by responders (77%) between conditions. However, this result does not mean that responders were not affected by context; as also shown in Table1 , they were responding to exchanges that were more helpful to them in the task condition. We expected this pattern to hold for accepted offers as well; thus, we expected that the offers that were accepted by responders were more helpful to them in the task condition than in the table condition.... ..."

"... In PAGE 10: ... Thus, accesses from programming lan- guages which were not supported before would be enabled, and, furthermore, additional operating systems and hardware platforms could be inte- grated. Table2 contains a summary of the pros and cons of the different approaches stated above. Since the BAPI/IDL approach seems to be the most benefi- cial, it will be the focus of attention in the following subsection of our paper.... ..."

"... In PAGE 8: ... Thus, accesses from programming lan- guages which were not supported before would be enabled, and, furthermore, additional operating systems and hardware platforms could be inte- grated. Table2 contains a summary of the pros and cons of the different approaches stated above. Since the BAPI/IDL approach seems to be the most bene- ficial, it will be the focus of attention in the fol- lowing subsection of our paper.... ..."

"... In PAGE 28: ... To gain a deeper insight into the relative effect of cognitive accuracy on performance across all five selling strategies, we conducted a regression analysis on the simulation data shown in Figures 3 and 4. Table2... In PAGE 29: ... We noted above that, as their cognitive accuracy increases, the agents become more self-confident and increase the concentration of the possible prices they could offer during negotiation around an average value. In the light of this, one way of interpreting the result in Table2 is that, the higher the degree of cognitive accuracy, the higher the minimum price that the seller is willing to accept from either the small investors or the active investor and, hence, the higher the price that may eventually be agreed upon. Thus, on average, as the seller apos;s cognitive representations become more accurate, the influence that he can exercise on his counter-parts becomes more effective in terms of a higher price at which the firm is sold.... In PAGE 29: ... This result refers to the average performance across the five selling strategies. However, when we look at the interaction effects between accuracy and selling strategy, Table2 complements our findings in three important respects. First, the benefits of an increase in accuracy are emphasised when strategy B* is used.... In PAGE 29: ... In contrast, as cognitive accuracy increases, the increase in revenue becomes lower if the seller negotiates simultaneously with the active investor and the small investors, or if only the small investors are involved in negotiation. Finally, Table2 shows that the impact of cognitive accuracy on effectiveness is negative when strategies A and B are used. Under these strategies, increasing the accuracy of the seller apos;s beliefs generates lower revenue.... In PAGE 30: ...-statistics. * = p lt; 0.001; ** = p lt; 0.01; *** = p lt; 0.05. an increase in cognitive accuracy to generate the most significant increase in revenue. When we look at the impact of cognitive accuracy on efficiency, Table2 shows that, on average, the more accurate the agents apos; beliefs are, the longer it takes to reach an agreement. As we noted above, the reason for this is that, as cognitive accuracy increases, the agents become more self- confident and, accordingly, more reluctant to offer prices that deviate from an average one.... In PAGE 30: ... To explore this trade-off, we now look at how cognitive accuracy and selling strategy interact to impact on efficiency. Table2... In PAGE 31: ... Whereas the former adjustment determines acceptance of a final lower price, the latter generates an increase in the number of messages that need to be exchanged before an agreement is reached. Finally, Table2 shows that when strategies C and D are used, the impact of accuracy on efficiency becomes positive. Under both strategies, increasing accuracy determines an improvement in efficiency.... ..."

"... In PAGE 6: ... To test this hypothesis, we performed a within-round com- parison of the offer benefit in both conditions. Table2 presents the number of rounds in which the difference between the proposed benefit for proposers and responders was positive (column Proposer gt; Responder ) and the number of rounds in which this dif- ference was negative (column Proposer lt; Responder ). As shown by the table, table proposers made offers that benefited themselves over responders significantly more of- ten than task proposers (chi-square p lt; 0.... In PAGE 6: ...en than task proposers (chi-square p lt; 0.05). These results confirm that table proposers are more likely to be competitive than proposers. Table2 also shows that 62% of all offers made by table proposers benefited them- selves more than table responders, while 60% of all offers made by task proposers ben- efited task responders more than themselves (chi-square p lt; 0.05).... ..."

"... In PAGE 6: ... To test this hypothesis, we performed a within-round com- parison of the offer benefit in both conditions. Table2 presents the number of rounds in which the difference between the proposed benefit for proposers and responders was positive (column Proposer gt; Responder ) and the number of rounds in which this dif- ference was negative (column Proposer lt; Responder ). As shown by the table, table proposers made offers that benefited themselves over responders significantly more of- ten than task proposers (chi-square p lt; 0.... In PAGE 6: ...en than task proposers (chi-square p lt; 0.05). These results confirm that table proposers are more likely to be competitive than proposers. Table2 also shows that 62% of all offers made by table proposers benefited them- selves more than table responders, while 60% of all offers made by task proposers ben- efited task responders more than themselves (chi-square p lt; 0.05).... ..."