diff --git a/papers/jaime_arias/figures/pyhrf_output.png b/papers/jaime_arias/figures/pyhrf_output.png
index 0860930a71..61628af4ed 100644
Binary files a/papers/jaime_arias/figures/pyhrf_output.png and b/papers/jaime_arias/figures/pyhrf_output.png differ
diff --git a/papers/jaime_arias/paper.rst b/papers/jaime_arias/paper.rst
index 2d51f8a88b..733f1109a7 100644
--- a/papers/jaime_arias/paper.rst
+++ b/papers/jaime_arias/paper.rst
@@ -591,7 +591,7 @@ the *left primary sensorimotor cortex*.
 
 .. figure:: figures/pyhrf_output.png
    :align: center
-   :scale: 35%
+   :scale: 27%
    :figclass: w
 
    PPMs (upper left), active parcels (lower left) and estimated HRFs (right),
@@ -613,7 +613,7 @@ the reliability of the results.
 
 PyHRF is an open source software, which has allowed it to evolve rapidly over
 the last few years. As we showed, it allows to generate posteriori probability
-maps (PPM) to describe the significance of the activation in each region of the
+maps (PPMs) to describe the significance of the activation in each region of the
 brain. Moreover, PyHRF uses efficient estimation methods in order to provide
 a fast and reliable tool. In fact, in 2013, a similar solution based on the
 BOLD JDE was developed in PyHRF for the Functional Arterial Spin Labelling