Traditionally, gambling is an activity involving two or more players in which the loser pays a bet to the winner and the outcome is uncertain. For example, play cards, mahjong, lottery, roulette and so on. Most people lose a large amount of money due to inappropriate gambling behavior, but however they continue to gamble to recoup previous losses and this phenomenon is named ‘chase’. Their wrong decisions are affected by poor gambling cognition, just like an illusion of control. In contrast, some people believe that people always win are more likely to become winner because he/she is lucky, and then he/she is more likely to bet again after winning, which is called ‘house money effect’.

An unreasonable phenomenon is that the same amount of money has a different impact on people's decision making, which could be explained by the notion of mental accounting. Money with different functions cannot be substituted, and money obtained with less effort is more likely to be consumed, especially in luxury.

A scale including five closely related parts is generally used to investigate people’s gambling cognitive bias. It is an open question that whether gambling behavior would change with situations when money is used for different places or not. We supposed that the ‘house money effect’ would get supported when people had a stronger tendency in gain situation, and the ‘chase’ would get supported when they had a stronger tendency in loss situation. However, it is unknown whether the two phenomena would appear in the same person or not.

The purpose of the current study was to explore the effectiveness of cognition and gambling fund on people’s gambling tendencies under the loss and gain situation. Three hundred and twenty-two participants (women 154) (M ± SD = 30.05 ± 11.97 years) were recruited to complete the online questionnaires, including the Gambling Related Cognition Scale (GRCS) and four self-designed items. A 2 (situation: loss, gain) × 2 (cognitive bias: high, low) × 2 (money: windfall gain, salary) analysis of variance was used to analyze the data. Results found that there were three significant main effects of situation, cognitive bias, and money, i.e., participants showed stronger gambling tendency to gain situation, and so had high cognitive bias and windfall gain. We used stepwise regression analysis and found that the different dimensions of cognitive bias played different roles in people’s gambling tendencies.

Participants in gain situation had a stronger gambling tendency, but there were no significant differences among some people who had participated in more than 4 kinds of gambling activities. There was a stronger gambling tendency when money used for gambling belonged to windfall gain. If participants had higher gambling cognitive biases, they were more obsessed with gambling plays in all situations. The different cognitive biases of gambling played different roles in people’s gambling tendencies. This study implies that we could reduce people’s gambling tendency by changing their cognition of gambling in case situation and money are fixed.