求助各位大神！
IRT理论的Graded Response Model中有一个描述Latent Trait Score的Extremity Parameters应该如何理解？看文献说的：
The extremity parameters show the latent trait score at which people have a 50/50 chance of selecting certain responses.

对于这个结果：

## Coefficients:
##          Extrmt1   Extrmt2    Extrmt3    Extrmt4     Extrmt5     Dscrmn
## A1     -1.790     1.230        2.974        5.211         8.695         0.407

有如下解释：

The Extrmt1 is -1.790 suggesting that those with a latent trait score of -1.790 have a 50/50 chance of selecting a 1 for that item. The Extrmt2 is 1.230 and so those with a latent trait score of 1.230 have a 50/50 chance of selecting a 1 or a 2 for that item. Extrmt3 is the latent trait score at which people have a 50/50 chance of selecting a 1, a 2 or a 3. Extrmt4 is the latent trait score at which people have a 50/50 chance of selecting a 1, a 2, a 3 or a 4. Finally, Extrmt5 is the latent trait score at which people have a 50/50 chance of selecting a 1, 2, 3, 4 or 5.

主要是这里的潜在特质得分和Extremity参数该如何理解和应用，不清楚。

请大神指点！